Novel segmented stacked autoencoder for effective dimensionality reduction and feature extraction in hyperspectral imaging

Convolutional neural network (CNN) is well known for its capability of feature learning and has made revolutionary achievements in many applications, such as scene recognition and target detection. In this paper, its capability of feature learning in hyperspectral images is explored by constructing a five-layer CNN for classification (C-CNN). The proposed C-CNN is constructed by including recent advances in deep learning area, such as batch normalization, dropout, and parametric rectified linear unit (PReLU) activation function. In addition, both spatial context and spectral information are elegantly integrated into the C-CNN such that spatial-spectral features are learned for hyperspectral images. A companion feature-learning CNN (FL-CNN) is constructed by extracting fully connected feature layers in this C-CNN. Both supervised and unsupervised modes are designed for the proposed FL-CNN to learn sensor-specific spatial-spectral features. Extensive experimental results on four benchmark data sets from two well-known hyperspectral sensors, namely airborne visible/infrared imaging spectrometer (AVIRIS) and reflective optics system imaging spectrometer (ROSIS) sensors, demonstrate that our proposed C-CNN outperforms the state-of-the-art CNN-based classification methods, and its corresponding FL-CNN is very effective to extract sensor-specific spatial-spectral features for hyperspectral applications under both supervised and unsupervised modes.

Learning Sensor-Specific Spatial-Spectral Features of Hyperspectral Images via Convolutional Neural Networks

Due to the remarkable achievements obtained by deep learning methods in the fields of computer vision, an increasing number of researches have been made to apply these powerful tools into hyperspectral image (HSI) classification. So far, most of these methods utilize a pre-training stage followed by a fine-tuning stage to extract deep features, which is not only tremendously time-consuming but also depends largely on a great deal of training data. In this study, we propose an efficient deep learning based method, namely, Random Patches Network (RPNet) for HSI classification, which directly regards the random patches taken from the image as the convolution kernels without any training. By combining both shallow and deep convolutional features, RPNet has the advantage of multi-scale, which possesses a better adaption for HSI classification, where different objects tend to have different scales. In the experiments, the proposed method and its two variants RandomNet and RPNet–single are tested on three benchmark hyperspectral data sets. The experimental results demonstrate the RPNet can yield a competitive performance compared with existing methods.

Hyperspectral image classification via a random patches network

Feature extraction is of high importance for effective data classification in hyperspectral imaging (HSI). Considering the high correlation among band images, spectral-domain feature extraction is widely employed. For effective spatial information extraction, a 2-D extension to singular spectrum analysis (2D-SSA), which is a recent technique for generic data mining and temporal signal analysis, is proposed. With 2D-SSA applied to HSI, each band image is decomposed into varying trends, oscillations, and noise. Using the trend and the selected oscillations as features, the reconstructed signal, with noise highly suppressed, becomes more robust and effective for data classification. Three publicly available data sets for HSI remote sensing data classification are used in our experiments. Comprehensive results using a support vector machine classifier have quantitatively evaluated the efficacy of the proposed approach. Benchmarked with several state-of-the-art methods including 2-D empirical mode decomposition (2D-EMD), it is found that our proposed 2D-SSA approach generates the best results in most cases. Unlike 2D-EMD that requires sequential transforms to obtain detailed decomposition, 2D-SSA extracts all components simultaneously. As a result, the execution time in feature extraction can be also dramatically reduced. The superiority in terms of enhanced discrimination ability from 2D-SSA is further validated when a relatively weak classifier, i.e., the k-nearest neighbor, is used for data classification. In addition, the combination of 2D-SSA with 1-D principal component analysis (2D-SSA-PCA) has generated the best results among several other approaches, demonstrating the great potential in combining 2D-SSA with other approaches for effective spatial-spectral feature extraction and dimension reduction in HSI.

/pdf/novel-two-dimensional-singular-spectrum-analysis-for-4g6nolcji7.pdf

Novel Two-Dimensional Singular Spectrum Analysis for Effective Feature Extraction and Data Classification in Hyperspectral Imaging

In recent years, deep neural networks have been successfully applied to model visual concepts and have achieved competitive performance on many tasks. Despite their impressive performance, traditional deep networks are subjected to the decayed performance under the condition of lacking sufficient training data. This problem becomes extremely severe for deep networks trained on a very small dataset, making them overfitting by capturing nonessential or noisy information in the training set. Toward this end, we propose a novel generalized deep transfer networks (DTNs), capable of transferring label information across heterogeneous domains, textual domain to visual domain. The proposed framework has the ability to adequately mitigate the problem of insufficient training images by bringing in rich labels from the textual domain. Specifically, to share the labels between two domains, we build parameter- and representation-shared layers. They are able to generate domain-specific and shared interdomain features, making this architecture flexible and powerful in capturing complex information from different domains jointly. To evaluate the proposed method, we release a new dataset extended from NUS-WIDE at http://imag.njust.edu.cn/NUS-WIDE-128.html. Experimental results on this dataset show the superior performance of the proposed DTNs compared to existing state-of-the-art methods.

Generalized Deep Transfer Networks for Knowledge Propagation in Heterogeneous Domains

As a widely used approach for feature extraction and data reduction, Principal Components Analysis (PCA) suffers from high computational cost, large memory requirement and low efficacy in dealing with large dimensional datasets such as Hyperspectral Imaging (HSI). Consequently, a novel Folded-PCA is proposed, where the spectral vector is folded into a matrix to allow the covariance matrix to be determined more efficiently. With this matrix-based representation, both global and local structures are extracted to provide additional information for data classification. Moreover, both the computational cost and the memory requirement have been significantly reduced. Using Support Vector Machine (SVM) for classification on two well-known HSI datasets and one Synthetic Aperture Radar (SAR) dataset in remote sensing, quantitative results are generated for objective evaluations. Comprehensive results have indicated that the proposed Folded-PCA approach not only outperforms the conventional PCA but also the baseline approach where the whole feature sets are used.

Novel Folded-PCA for improved feature extraction and data reduction with hyperspectral imaging and SAR in remote sensing

Gradient-based subspace phase correlation for fast and effective image alignment

Maximum likelihood classifier (MLC) and support vector machines (SVM) are two commonly used approaches in machine learning. MLC is based on Bayesian theory in estimating parameters of a probabilistic model, whilst SVM is an optimization based nonparametric method in this context. Recently, it is found that SVM in some cases is equivalent to MLC in probabilistically modeling the learning process. In this paper, MLC and SVM are combined in learning and classification, which helps to yield probabilistic output for SVM and facilitate soft decision making. In total four groups of data are used for evaluations, covering sonar, vehicle, breast cancer, and DNA sequences.The data samples are characterized in terms of Gaussian/non-Gaussian distributed and balanced/unbalanced samples which are then further used for performance assessment in comparing the SVM and the combined SVM-MLC classifier. Interesting results are reported to indicate how the combined classifier may work under various conditions.

/pdf/combining-mlc-and-svm-classifiers-for-learning-based-2bpv141vrh.pdf

Yi Zhang

Papers

Novel Folded-PCA for improved feature extraction and data reduction with hyperspectral imaging and SAR in remote sensing

Gradient-based subspace phase correlation for fast and effective image alignment

Combining MLC and SVM classifiers for learning based decision making: analysis and evaluations