A new digital repository for remotely sensed hyperspectral imagery with unmixing-based retrieval functionality

Hyperspectral images contain information from a wider range of the electromagnetic spectrum than natural images which gives them potential for better classification ability. However, hyperspectral datasets are typically small due to the expensive equipment needed to obtain the images, which can limit classification performance. One solution to this problem is transfer learning, in which a model trained on one dataset is reused for a separate dataset. Research has shown that transfer learning between hyperspectral datasets can give improved performance over models without transfer learning when training data are limited. Since extra hyperspectral data are not always available, the solution proposed here is to instead use networks pretrained on natural image (i.e., red, blue, green, or RGB) datasets for transfer learning. By using various feature selection and feature extraction methods, extracted hyperspectral samples are transformed into a three-channel format to imitate an RGB image and are used for fine tuning the well-known ResNet, DenseNet, and VGG networks. Feature extraction methods include techniques like principal component analysis, which create lower dimensional features from high dimensional spectral data. Alternatively, feature selection methods aim to find the best set of existing channels to use for classification. Experimental results are obtained using two well-known hyperspectral datasets, showing 73.6% accuracy on Pavia University and 82.8% accuracy on Salinas with 25 training samples per class. Additional ensemble methods are implemented that utilize multiple networks and show an increase in accuracy of 4.4% and 3% for Pavia University and Salinas, respectively. These results demonstrate that networks pretrained on RGB datasets are suitable for transfer learning with hyperspectral image datasets and can achieve desirable performance given the proper preprocessing technique.

The scarcity of labeled training samples poses a significant challenge in hyperspectral image classification. Cross-scene classification has been shown to be an effective approach to tackle the problem of limited sample learning. This paper investigates the usage of generative adversarial networks (GANs) to enable collaborative artificial intelligence learning on hyperspectral datasets. We propose and design a specialized architecture, termed Hyper-CycleGAN, for heterogeneous transfer learning across source and target scenes. This architecture enables the establishment of bidirectional mappings through efficient adversarial training and merges both source-to-target and target-to-source generators. The proposed Hyper-CycleGAN architecture harnesses the strengths of GANs, along with custom modifications like the integration of multi-scale attention mechanisms to enhance feature learning capabilities specifically tailored for hyperspectral data. To address training instability, the Wasserstein generative adversarial network with gradient penalty (WGAN-GP) loss discriminator is utilized. Additionally, a label smoothing technique is introduced to enhance the generalization capability of the generator, particularly in handling unlabeled samples, thus improving model robustness. Experimental results are performed to validate and confirm the effectiveness of the cross-domain Hyper-CycleGAN approach by demonstrating its applicability to two real-world cross-scene hyperspectral image datasets. Addressing the challenge of limited labeled samples in hyperspectral image classification, this research makes significant contributions and gives valuable insights for remote sensing, environmental monitoring, and medical imaging applications.

In this paper, a contextual Support Vector Machine (SVM) technique based on the principle of Hilbert Space Embedding (HSE) of a local hyperspectral data distribution into an Reproducing Kernel Hilbert Space (RKHS) is proposed to optimally exploit the spectral and local spatial information of the hyperspectral image. The idea of embedding is to map hyperspectral pixels in a local neighborhood into a single point in the RKHS that can uniquely represent those pixels collectively. Previously, the authors have employed an HSE called empirical mean map to build the contextual SVM. In this work, a weighted empirical mean map is utilized to exploit the similarities and variation in the local spatial information. For every pixel, a small set of the neighboring pixels in a hyperspectral image are mapped into an RKHS induced by a certain kernel (Eg. Gaussian RBF kernel) and then, the embedded point of these group of pixels is obtained by calculating the weighted empirical mean of these mapped points. The weights are determined based on the distance between the pixel in consideration and its neighbors. An SVM separating hyperplane is built to maximize the margin between classes formed by weighted empirical means. The proposed technique showed significant improvement over the existing contextual and composite kernels on two hyperspectral image data sets.

The use of matched filters on hyperspectral data has made it possible to detect faint signatures. This study uses a modified k-means clustering to improve matched filter performance. Several simple bivariate cases are examined in detail, and the interaction of filtering and partitioning is discussed. The authors show that clustering can reduce within-class variance and group pixels with similar correlation structures. Both of these features improve filter performance. The traditional k-means algorithm is modified to work with a sample of the image at each iteration and is tested against two hyperspectral datasets. A new "extreme" centroid initialization technique is introduced and shown to speed convergence. Several matched filtering formulations (the simple matched filter, the clutter matched filter, and the saturated matched filter) are compared for a variety of number of classes and synthetic hyperspectral images. The performance of the various clutter matched filter formulations is similar, all are about an order of magnitude better than the simple matched filter. Clustering is found to improve the performance of all matched filter formulations by a factor of two to five. Clustering in conjunction with clutter matched filtering can improve fifty-fold over the simple case, enabling very weak signals to be detected in hyperspectral images.

In hyperspectral images (HSI) classification, it is important to combine multiple features of a certain pixel in both spatial and spectral domains to improve the classification accuracy. To achieve this goal, this article proposes a novel spatial-spectral feature dimensionality reduction algorithm based on manifold learning. For each feature, a graph Laplacian matrix is constructed based on discriminative information from training samples, and then the graph Laplacian matrices of the various features are linearly combined using a set of empirically defined weights. Finally, the feature mapping is obtained by an eigen-decomposition problem. Based on the classification results of the public Indiana Airborne Visible Infrared Imaging Spectrometer dataset and Texas Hyperspectral Digital Imagery Collection Experiment data set, the technical accuracies show that our method achieves superior performance compared to some representative HSI feature extraction and dimensionality reduction algorithms.

Hyperspectral (HS) image is a three dimensional data image where the 3rd dimension carries the wealth of spectrum information. HS image compression is one of the areas that has attracted increasing attention for big data processing and analysis. HS data has its own distinguishing feature which differs with video because without motion, also different with a still image because of redundancy along the wavelength axis. The prediction based method is playing an important role in the compression and research area. Reflectance distribution of HS based on our analysis indicates that there is some nonlinear relationship in intra-band. The Multilayer Propagation Neural Networks (MLPNN) with backpropagation training are particularly well suited for addressing the approximation function. In this paper, an MLPNN based predictive image compression method is presented. We propose a hybrid Adaptive Prediction Mechanism (APM) with MLPNN model (APM-MLPNN). MLPNN is trained to predict the succeeding bands by using current band information. The purpose is to explore whether MLPNN can provide better image compression results in HS images. Besides, it uses less computation cost than a deep learning model so we can easily validate the model. We encoded the weights vector and the bias vector of MLPNN as well as the residuals. That is the only few bytes it then sends to the decoder side. The decoder will reconstruct a band by using the same structure of the network. We call it an MLPNN decoder. The MLPNN decoder does not need to be trained as the weights and biases have already been transmitted. We can easily reconstruct the succeeding bands by the MLPNN decoder. APM constrained the correction offset between the succeeding band and the current spectral band in order to prevent HS image being affected by large predictive biases. The performance of the proposed algorithm is verified by several HS images from Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) reflectance dataset. MLPNN simulation results can improve prediction accuracy; reduce residual of intra-band with high compression ratio and relatively lower bitrates.

In this paper, we present new sparsity based algorithms in generating a high resolution hyperspectral image by fusing a high resolution color image with a low resolution hyperspectral image. Mathematical formulation of the sparsity based approaches is presented. Comparison with other pansharpening algorithms using actual data has been carried out using two hyperspectral image data sets. Initial results are encouraging. Most importantly, the new sparsity formulation points to a new direction in generating high resolution hyperspectral images where the raw images may be noisy.

This study explores the utilization of Binary Embedded Zero Tree Wavelet Algorithms (BEZW) to compress hyperspectral images. The primary goal is to enhance the representation of spectral data while minimizing storage and transmission requirements. The BEZW algorithm employs wavelet transformations to leverage both spectral and spatial redundancies found in hyperspectral data cubes. Its embedded zero tree structure ensures efficient encoding of small coefficients with minimal overhead. The study evaluates the performance of the BEZW method utilizing various hyper-spectral datasets and compression settings, comparing it to other compression techniques. The results illustrate that the BEZW algorithm offers a promising approach to hyperspectral image compression, achieving competitive compression ratios while preserving spectral accuracy. This makes it a valuable option for applications where efficient hyperspectral data storage and transmission are crucial. The research contributes to the field of hyperspectral imaging by introducing an effective compression method that enhances data accessibility, simplifying the utilization of hyperspectral data in resource-constrained environments.

ABSTRACTIn this article, we propose a method for extracting spatio-spectral features from high spatial resolution hyperspectral (HS) images. The method is based on extracting two-dimensional moments from neighbourhoods of pixels. Three different types of moments are considered: geometric, complex Zernike and Legendre. Moments of a given type are extracted from a few principal components (PC) of HS data, and are stacked on the original HS data to form a joint spatio-spectral feature space. These features are classified using a support vector machine (SVM) classifier. The influence of the moments orders and the size of the neighbourhood window on the quality of the extracted features are analysed. A few experiments are conducted on two widely used HS data sets, Pavia University and Salinas. The results demonstrate high capabilities of the proposed method in comparison with some state-of-the-art spatio-spectral HS classification methods.

This paper addresses a conceptual approach to hyperspectral image assessment that uses physics-based models to constrain multiparameter inversion algorithms aimed at quantitative measurement of material properties. Sensing approaches include use of a single hyperspectral image with many spectral samples, a sequence of images acquired over time, and widely different types of measurements from a range of sensor types. We propose a conceptual approach for merging these data into a common framework to allow simultaneous exploitation of these multiparameter data sets. Physics-based models predict or constrain the range of observable parameters associated with a target or material condition, then model matching or optimization methods invert a measurement set to a target type or condition. 2 examples are presented using hyperspectral data sets. The first involves characterizing atmospheric constituents using the MODTRAN Code to constrain the solutions. The second uses a radiation propagation model to drive an inversion of hyperspectral data to multiple water quality parameters. Finally, we discuss how more involved 3D physics-based synthetic image models may hold a key to image exploitation algorithms.

Cloud computing offers the possibility to store and process massive amounts of remotely sensed hyperspectral data in a distributed way. Dimensionality reduction is an important task in hyperspectral imaging, as hyperspectral data often contains redundancy that can be removed prior to analysis of the data in repositories. In this regard, the development of dimensionality reduction techniques in cloud computing environments can provide both efficient storage and preprocessing of the data. In this paper, we develop a parallel and distributed implementation of a widely used technique for hyperspectral dimensionality reduction: principal component analysis (PCA), based on cloud computing architectures. Our implementation utilizes Hadoop's distributed file system (HDFS) to realize distributed storage, uses Apache Spark as the computing engine, and is developed based on the map-reduce parallel model, taking full advantage of the high throughput access and high performance distributed computing capabilities of cloud computing environments. We first optimized the traditional PCA algorithm to be well suited for parallel and distributed computing, and then we implemented it on a real cloud computing architecture. Our experimental results, conducted using several hyperspectral datasets, reveal very high performance for the proposed distributed parallel method.

In this study, we demonstrate the potential of a non-destructive hyperspectral imaging processing method in the near-infrared (NIR) region (874–1734 nm) for classifying the quality of brined kimchi cabbage. The salinity level of brined kimchi cabbage is an important indicator of consumer preference and the quality of kimchi. Hence, we compared the water content and salinity of brined kimchi cabbage via hyperspectral data. We extracted the optimal wavelengths from the hyperspectral image dataset to classify the salinity level of the predicted brined kimchi cabbage, and thus, established a novel approach for classifying kimchi samples into quality-unacceptable and quality-acceptable groups. Standard normal variate and multiplicative scatter correction (MSC) were used for pathlength correction. The Savitzky–Golay first and second derivatives were used for the deconvolution of the raw spectral data. The experimental results confirmed that the decision tree model combined with MSC pathlength correction and Savitzky–Golay first derivative preprocessing was the best classification model. The proposed hyperspectral image–NIR system can be applied to the detection of salinity during industrial kimchi manufacturing.

Development in the field of computer-aided learning and testing have stimulated the progress of novel and efficient knowledge-based expert systems that have shown hopeful outcomes in a broad variety of practical applications. In particular, deep learning techniques have been extensively carried out to identify remote sensed data obtained by the instruments of Earth observation. Hyperspectral imaging (HSI) is an evolving area in the study of remotely sensed data due to the huge volume of information found in these images, which enables better classification and processing of the Earth’s surface by integrating ample of spatial and spectral features. Nevertheless, because of the high-dimensional data and restricted training samples available, HSI presents some crucial challenges for classification of supervised methods. In particular, it addresses the problems of spectral and spatial resolution, volume of data, and model conversion from multimedia images to hyperspectral data. Various deep learning-based architectures are currently being established to solve these limitations, showing significant results in the analysis of hyperspectral data. In this paper, we deal primarily with the hyperspectral datasets, the dimensionality curse problem, and methods for classifying those datasets using some deep neural networks (DNN), especially convolutional neural networks (CNN). We provide a comparative analysis of various dimensionality reduction (DR) and classification techniques used for finding accuracies based on the datasets used. We also explore certain hyperspectral imaging applications along with some of the research axes.

The performances of regular support vector machines and random forests are experimentally compared for hyperspectral imaging land cover classification. Special characteristics of hyperspectral imaging dataset present diverse processing problems to be resolved under robust mathematical formalisms such as image classification. As a result, pixel purity index algorithm is used to obtain endmember spectral responses from Indiana pine hyperspectral image dataset. The generalized reduced gradient optimization algorithm is thereafter executed on the research data to estimate fractional abundances in the hyperspectral image and thereby obtain the numeric values for land cover classification. The Waikato environment for knowledge analysis (WEKA) data mining framework is selected as a tool to carry out the classification process by using support vector machines and random forests classifiers. Results show that performance of support vector machines is comparable to that of random forests. This study makes a positive contribution to the problem of land cover classification by exploring generalized reduced gradient method, support vector machines, and random forests to improve producer accuracy and overall classification accuracy. The performance comparison of these classifiers is valuable for a decision maker to consider tradeoffs in method accuracy versus method complexity.

Recent growth in unmanned aerial vehicle (UAV) technology have promoted the detailed mapping of individual tree species. However, the in-depth mining and comprehending of the significance of features derived from high-resolution UAV data for tree species discrimination remains a difficult task. In this study, a state-of-the-art approach combining UAV-borne light detection and ranging (LiDAR) and hyperspectral was used to classify 11 common tree species in a typical natural secondary forest in Northeast China. First, comprehensive relevant structural and spectral features were extracted. Then, the most valuable feature sets were selected by using a hybrid approach combining correlation-based feature selection with the optimized recursive feature elimination algorithm. The random forest algorithm was used to assess feature importance and perform the classification. Finally, the robustness of features derived from point clouds with different structures and hyperspectral images with different spatial resolutions was tested. Our results showed that the best classification accuracy was obtained by combining LiDAR and hyperspectral data (75.7%) compared to that based on LiDAR (60.0%) and hyperspectral (64.8%) data alone. The mean intensity of single returns and the visible atmospherically resistant index for red-edge band were the most influential LiDAR and hyperspectral derived features, respectively. The selected features were robust in point clouds with a density not lower than 5% (~5 pts/m2) and a resolution not lower than 0.3 m in hyperspectral data. Although canopy surface features were slightly different from original LiDAR features, canopy surface information was also important for tree species classification. This study proved the capabilities of UAV-borne LiDAR and hyperspectral data in natural secondary forest tree species discrimination and the potential for this approach to be transferable to other study areas.