The Deep Belief and Self-Organizing Neural Network as a Semi-Supervised Classification Method for Hyperspectral Data

Abstract

Hyperspectral data is not linearly separable, and it has a high characteristic dimension. This paper proposes a new algorithm that combines a deep belief network based on the Boltzmann machine with a self-organizing neural network. The primary features of the hyperspectral image are extracted with a deep belief network. The weights of the network are fine-tuned using the labeled sample. Feature vectors extracted by the deep belief network are classified by a self-organizing neural network. The method reduces the spectral dimension of the data while preserving the large amount of original information in the data. The method overcomes the long training time required when using self-organizing neural networks for clustering, as well as the training difficulties of Deep Belief Networks (DBN) when the labeled sample size is small, thereby improving the accuracy and robustness of the semi-supervised classification. Simulation results show that the structure of the network can achieve higher classification accuracy when the labeled sample is deficient.

Sparse Representation Graph for Hyperspectral Image Classification Assisted by Class Adjusted Spatial Distance

“…However, among all HSI data acquired, the labeled one is very limited. In this situation, semi-supervised learning (SSL) provides a promising way to deal with both the limited labeled data and the rich unlabeled data [6,7].…”

Fisher Discriminative Sparse Representation Based on DBN for Fault Diagnosis of Complex System

“…To reveal the deeper connection between the variables and create a proper representation of physical meaning, this paper proposes a novel feature extraction scheme for complex system fault diagnosis using deep learning and sparse representation. Deep learning allows computational models composed of multiple processing layers to learn the representation of data [18][19][20][21]. The backpropagation algorithm (BP algorithm) is applied after the weights training step in deep learning to discover intricate structures in large data sets, and the BP algorithm makes up the disadvantage of gradient diffusion in pretraining weights of the network.…”