Open Access Open Access  Restricted Access Subscription Access

A Methodology for Unsupervised Feature Learning in Hyperspectral Imagery Using Deep Belief Network


Affiliations
1 Department of Electronics and Communication Engineering, Sri Venkateswara College of Engineering, Sriperumbudur 602 117, India
 

Deep learning approaches have received major interest in the field of remote sensing. Hyperspectral imaging has rich data that are distributed in multi-dimensions. It is challenging to apply deep learning algorithms due to the limited amount of labelled data. So, unsupervised feature extraction approaches are used to overcome this limitation. In this study, we propose an unsupervised feature learning approach using deep belief network (DBN). In the proposed framework, the input hyperspectral image is segmented using entropy rate superpixel segmentation and filtered by domain transform recursive filter which extracts spatial and spectral information effectively. Then the features are learned by improved DBN. In the traditional methods, DBN is stacked with restricted Boltzmann machine (RBM) which is suitable for only binary value data. In the proposed framework, we used Gaussian–Bernoulli RBM which was constructed for real value data such as images. The experiments were carried out using Pavia University dataset. The results show that the proposed network has good performance in terms of classification accuracy and computation time.

Keywords

Deep Belief Network, Hyperspectral Image, Remote Sensing, Spatial–Spectral Classification, Superpixel Segmentation.
User
Notifications
Font Size


  • A Methodology for Unsupervised Feature Learning in Hyperspectral Imagery Using Deep Belief Network

Abstract Views: 436  |  PDF Views: 185

Authors

C. Sherin Shibi
Department of Electronics and Communication Engineering, Sri Venkateswara College of Engineering, Sriperumbudur 602 117, India
R. Gayathri
Department of Electronics and Communication Engineering, Sri Venkateswara College of Engineering, Sriperumbudur 602 117, India

Abstract


Deep learning approaches have received major interest in the field of remote sensing. Hyperspectral imaging has rich data that are distributed in multi-dimensions. It is challenging to apply deep learning algorithms due to the limited amount of labelled data. So, unsupervised feature extraction approaches are used to overcome this limitation. In this study, we propose an unsupervised feature learning approach using deep belief network (DBN). In the proposed framework, the input hyperspectral image is segmented using entropy rate superpixel segmentation and filtered by domain transform recursive filter which extracts spatial and spectral information effectively. Then the features are learned by improved DBN. In the traditional methods, DBN is stacked with restricted Boltzmann machine (RBM) which is suitable for only binary value data. In the proposed framework, we used Gaussian–Bernoulli RBM which was constructed for real value data such as images. The experiments were carried out using Pavia University dataset. The results show that the proposed network has good performance in terms of classification accuracy and computation time.

Keywords


Deep Belief Network, Hyperspectral Image, Remote Sensing, Spatial–Spectral Classification, Superpixel Segmentation.

References





DOI: https://doi.org/10.18520/cs%2Fv120%2Fi11%2F1705-1711