Subset Of Bands Research Articles

Unsupervised Rate Distortion Function-Based Band Subset Selection for Hyperspectral Image Classification

Unsupervised Rate Distortion Function-Based Band Subset Selection for Hyperspectral Image Classification

Hyperspectral Image Band Selection Based on CNN Embedded GA (CNNeGA)

Hyperspectral images (HSI) are a powerful source of reliable data in various remote sensing applications. But due to the large number of bands, HSI has information redundancy, and methods are often used to reduce the number of spectral bands. Band selection (BS) is used as a preprocessing solution to reduce data volume, increase processing speed, and improve methodology accuracy. However, most conventional BS approaches are unable to fully explain the interaction between spectral bands and evaluate the representation and redundancy of the selected band subset. This study first examines a supervised band selection method that allows the selection of the required number of bands. A deep network with 3D-convolutional layers embedded in a genetic algorithm (GA) The GA uses embedded 3D-CNN (CNNeGA) as a fitness function. GA also considers the parent check box. The parent check box (parent subbands) is designed to make genetic operators more effective. In addition, the effectiveness of increasing the attention layer to a 3D-CNN and converting this model to spike neural networks (SNNs) has been investigated in terms of accuracy and complexity over time. The evaluation of the proposed method and the obtained results are satisfactory. The accuracy improved from 6 to 21 percent. Accuracy between 90 to 99 percent has been obtained in each evaluation mode.

Hyperspectral imaging with a band matrix reduction method to detect early drought stress in tomato

Drought stress is one of the key abiotic stresses affecting plant growth, crop yield and food quality. The main objective of this study is to investigate the potential effectiveness of hyperspectral imaging with band selection method for the rapid detection of the early drought stress of tomatoes. First, the unsupervised algorithm - K-means and statistical histogram are used to extract samples representing each experimental treatment group. Then, to solve problems related to the high redundancy and correlation of hyperspectral data, band matrix reduction method (BMRM) based on recursive feature elimination theory is proposed to determine the optimal band subset. The band matrix is constructed according to the band ranking obtained by the discrimination coefficient -Coefi, which is calculated from the average spectral curve and the first-derivative spectrum. Finally, the effectiveness of waveband selection algorithms was validated by comparison with successive projections algorithm, competitive adaptive reweighted sampling, recursive feature elimination with cross-validation and full spectrum. The results demonstrated that BMRM achieved higher classification accuracy with fewer bands selected, and the amount of calculation is not greatly improved. The proposed method provides a more accurate, and effective way of detecting early drought stress.

A Multitask Bee Colony Band Selection Algorithm With Variable-Size Clustering for Hyperspectral Images

Band selection (BS) is a widely used dimensionality reduction technique for hyperspectral images. However, most of existing evolutionary algorithms focus on searching a globally optimal band subset under a fixed size, and their obtained band subsets may still contain a large number of redundant bands. In order to simultaneously obtain multiple optimal band subsets with different sizes, this article proposes an unsupervised multitask artificial bee colony (ABC) BS algorithm based on variable-size clustering (MBBS-VC). First, a variable-size band clustering method based on worst class decomposition is developed, based on which the BS problem can be modeled as a multitask optimization problem. Next, a multitask multimicrogroup bee colony algorithm with variable coding length is proposed to simultaneously search multiple optimal band subsets with different sizes. Moreover, several new strategies, including the intergroup collaboration strategy based on bidirectional neighborhood learning and the multimeasure integration judgment (MIJ) mechanism, are designed to improve the performance of MBBS-VC. In this article, the hyperspectral BS problem is transformed into a multitask optimization problem for the first time. Finally, compared with 15 classical BS algorithms on several commonly used datasets, experimental results verify the superiority of the proposed BS algorithm.

Retrieval of crop biophysical-biochemical variables from airborne AVIRIS-NG data using hybrid inversion of PROSAIL-D

Sustainable agricultural growth and management reduces over-utilization of farm resources and squeezes risk of negative impacts on environment. Monitoring continuous crop growth and health under various conditions at different spatio-temporal resolutions is a key to assess yield stability, crop diversity, adaptability, mitigation for stress and response. The quantification of crop biophysical and biochemical variables spatially from remotely sensed data with help of spectroscopic methods provide a reliable discerning information in the context of crop foliar condition like leaf greenness, senescence, canopy density, crop growth, stress and eco-physiological processes. Airborne Visible/Infrared Imaging Spectrometer-Next Generation (AVIRIS-NG) airborne hyperspectral data offers high spatial and spectral resolution giving a unique advantage and opportunity to test retrievals of crop biophysical (BP)-biochemical (BC) variables under varying conditions over different types of crops. A hybrid inversion of leaf-canopy Radiative Transfer model – PROSAIL-D complemented with use of data driven nonlinear non-parametric methods which offer simplicity, fastness, reliability and competency is a powerful method to retrieve crop biophysical and biochemical variables. The Hyperspectral band (feature) selection is a computationally cost efficient method to overcome data redundancy in high dimensional correlated input spectral bands. The determination of optimum subset or combination of hyperspectral bands specific to retrieval of vegetation properties (including canopy effects) determined using feature selection algorithm for regression-based retrievals are active research topic unlike classification problems in which it is more common. A band selection algorithm based on Gaussian Processes Regression was used to choose most sensitive bands from in-situ biophysical-biochemical measurements and crop spectral signatures collected for analysis in two different agricultural regions: Raichur (Karnataka) and Anand (Gujarat) districts of India representing diverse landscapes, heterogeneous crop canopies and agro climatic settings. The retrieval algorithm for AVIRIS-NG image employed a decision tree ensemble algorithm Canonical Correlation Forests using the optimum subset of bands for retrieval of targeted crop variable. Validation of retrieved crop variables were done using in-situ ground observations collected over heterogeneous diverse crop landscape. The results showed chlorophyll (RMSE = 6.61 µg cm−2), equivalent water thickness (RMSE = 0.002 cm), leaf area index (RMSE = 0.35 m2/m−2) and dry matter (RMSE = 0.003 g cm−2), carotenoid (RMSE = 14.3 µg cm−2), anthocyanin (RMSE = 12.92 µg cm−2) retrieved with better accuracies. The results obtained in context of feature (band) selection approach for Radiative Transfer model inversion show overlapping of sensitive spectral bands of chlorophyll-ab, carotenoid and anthocyanin especially between narrow spectral range 480 to 560 nm as predominant reason for decreased accuracies of anthocyanin and carotenoid compared to other variable. This poses a limitation as well as opportunity for further research in signal separation in context of feature selection approach especially in context of broader spectral bandwidths. It also sets ground for further development of feature selection algorithms that use hybrid regression methods customized for crop specific traits.

A Band Subset Selection Approach Based on Sparse Self-Representation and Band Grouping for Hyperspectral Image Classification

Band subset selection (BSS) is one of the ways to implement band selection (BS) for a hyperspectral image (HSI). Different from conventional BS methods, which select bands one by one, BSS selects a band subset each time and preserves the best one from the collection of the band subsets. This paper proposes a BSS method, called band grouping-based sparse self-representation BSS (BG-SSRBSS), for hyperspectral image classification. It formulates BS as a sparse self-representation (SSR) problem in which the entire bands can be represented by a set of informatively complementary bands. The BG-SSRBSS consists of two steps. To tackle the issue of selecting redundant bands, it first applies band grouping (BG) techniques to pre-group the entire bands to form multiple band groups, and then performs band group subset selection (BGSS) to find the optimal band group subset. The corresponding representative bands are taken as the BS result. To efficiently find the nearly global optimal subset among all possible band group subsets, sequential and successive iterative search algorithms are adopted. Land cover classification experiments conducted on three real HSI datasets show that BG-SSRBSS can improve classification accuracy by 4–20% compared to the existing BSS methods and requires less computation time.

Multiple Band Prioritization Criteria-Based Band Selection for Hyperspectral Imagery

Band selection (BS) is an effective pre-processing way to reduce the redundancy of hyperspectral data. Specifically, the band prioritization (BP) criterion plays an essential role since it can judge the importance of bands from a particular perspective. However, most of the existing methods select bands according to a single criterion, leading to incomplete band evaluation and insufficient generalization against different data sets. To address this problem, this work proposes a multi-criteria-based band selection (MCBS) framework, which innovatively treats BS as a multi-criteria decision-making (MCDM) problem. First, a decision matrix is constructed based on several typical BPs, so as to evaluate the bands from different focuses. Then, MCBS defines the global positive and negative idea solutions and selects bands according to their relative closeness to these solutions. Since each BP has a different capability to discriminate the bands, two weight estimation approaches are developed to adaptively balance the contributions of various criteria. Finally, this work also provides an extended version of MCBS, which incorporates the subspace partition strategy to reduce the correlation of the selected bands. In this paper, the classification task is used to evaluate the performance of the selected band subsets. Extensive experiments on three public data sets verify that the proposed method outperforms other state-of-the-art methods.

Electronic Spectroscopy of SmO in the 645 to 670 nm Range.

The associative ionization reaction Sm + O → SmO+ + e- is being investigated as an electron source that could transiently modify high-altitude electron densities via Sm vapor release. Electronic spectra have been obtained from tests where sounding rockets released Sm vapor, but the interpretation of these results has been hampered by the limited laboratory spectral data available for both SmO and SmO+. The present study extends the spectroscopic characterization of SmO in the 645-670 nm range, where the field data show the most prominent molecular emission features. Rotationally resolved excitation spectra, dispersed laser-induced fluorescence spectra, and fluorescence decay lifetimes are reported. The results are consistent with the assignment of a subset of the red-region bands to configurational transitions of the form Sm2+(4f56s)O2- ↔ Sm2+(4f55d)O2-. Analysis of the excited state hyperfine structure supports this configurational description.

Band selection for heterogeneity classification of hyperspectral transmission images based on multi-criteria ranking

Hyperspectral images contain hundreds of continuous spectral bands, which can provide rich information for target detection and image classification. However, high-dimensional data are easy to cause the “Hughes phenomenon”, which affects the classification accuracy of images. In this paper, a band selection method based on multi-criteria ranking is proposed for hyperspectral dimensionality reduction. First, the hyperspectral image acquisition system is built to collect the hyperspectral transmission image datasets containing various heterogeneities. Second, the original data are preprocessed and the band selection method proposed is used for data dimensionality reduction. That is, (1) the whole band is divided into subintervals according to the correlation criteria, and then the bands with large information amount and good inter-class separability are selected according to other criteria functions, (2) the bands of each subinterval are combined, and the combined bands are evaluated according to the optimum index factor and Bhattacharyya distance, and (3) the combined bands are ranked and the top-ranked combined bands are selected as the representative bands for image classification. Finally, the support vector machine is used to verify the effectiveness of the representative bands selected by this method. The experimental results show that the proposed method achieves data dimensionality reduction, and the overall classification accuracy can reach 100% on some selected subsets of bands, which has better classification performance than other related band selection methods. In addition, the computational complexity of high-dimensional data can be effectively reduced by band selection, and the average running time is 95.19% less than that of the full bands while ensuring classification accuracy.

Stochastic gate-based autoencoder for unsupervised hyperspectral band selection

Due to its strong feature representation ability, the deep learning (DL)-based method is preferable for the unsupervised band selection task of hyperspectral image (HSI). However, the current DL-based UBS methods have not further investigated the nonlinear relationship between spectral bands, a more robust DL model with effective loss function is desired. To solve the above problem, a novel stochastic gate-based autoencoder (SGAE) has been proposed for the UBS task. With the proposed stochastic gate layer, the desired band subset with learnable parameters can be directly obtained. For obtaining better UBS results, a nonlinear regularization term is added with the loss function to supervise the training process of SGAE. Furthermore, an early stopping criteria with a regularization term-based threshold is developed. Experimental results on four publicly available remote sensing datasets prove the effectiveness of our SGAE.

Optimal estimation framework for ocean color atmospheric correction and pixel-level uncertainty quantification.

Ocean color (OC) remote sensing requires compensation for atmospheric scattering and absorption (aerosol, Rayleigh, and trace gases), referred to as atmospheric correction (AC). AC allows inference of parameters such as spectrally resolved remote sensing reflectance (Rrs(λ);sr-1) at the ocean surface from the top-of-atmosphere reflectance. Often the uncertainty of this process is not fully explored. Bayesian inference techniques provide a simultaneous AC and uncertainty assessment via a full posterior distribution of the relevant variables, given the prior distribution of those variables and the radiative transfer (RT) likelihood function. Given uncertainties in the algorithm inputs, the Bayesian framework enables better constraints on the AC process by using the complete spectral information compared to traditional approaches that use only a subset of bands for AC. This paper investigates a Bayesian inference research method (optimal estimation [OE]) for OC AC by simultaneously retrieving atmospheric and ocean properties using all visible and near-infrared spectral bands. The OE algorithm analytically approximates the posterior distribution of parameters based on normality assumptions and provides a potentially viable operational algorithm with a reduced computational expense. We developed a neural network RT forward model look-up table-based emulator to increase algorithm efficiency further and thus speed up the likelihood computations. We then applied the OE algorithm to synthetic data and observations from the moderate resolution imaging spectroradiometer (MODIS) on NASA's Aqua spacecraft. We compared the Rrs(λ) retrieval and its uncertainty estimates from the OE method with in-situ validation data from the SeaWiFS bio-optical archive and storage system (SeaBASS) and aerosol robotic network for ocean color (AERONET-OC) datasets. The OE algorithm improved Rrs(λ) estimates relative to the NASA standard operational algorithm by improving all statistical metrics at 443, 555, and 667 nm. Unphysical negative Rrs(λ), which often appears in complex water conditions, was reduced by a factor of 3. The OE-derived pixel-level Rrs(λ) uncertainty estimates were also assessed relative to in-situ data and were shown to have skill.

Feature Extraction of Hyperspectral Images Based on Subspace Band Selection and Transform-Domain Recursive Filtering

During the feature extraction of hyperspectral images, a single filter cannot acquire complete information. To solve the problem, this paper proposes a feature extraction method based on subspace band selection and transform-domain recursive filtering. The proposed method contains three steps: Firstly, the target hyperspectral image is divided into multiple subsets of adjacent bands. Secondly, the Lasso-based band selection approach is adopted to compute the sparsity coefficient of each band. The bands in each subset are then ranked by the coefficient. Based on the ranking, the band with the highest coefficient is extracted from each subset, and used to reconstruct the hyperspectral data. Finally, the reconstructed hyperspectral image is processed through transform-domain recursive filtering, producing the features to be classified. Taking the support vector machine (SVM) as the classifier, our method was tested on several real hyperspectral image datasets. The results show that our method has a better classification accuracy than the other band selection methods.

Hyperspectral Band Selection via Optimal Combination Strategy

Band selection is one of the main methods of reducing the number of dimensions in a hyperspectral image. Recently, various methods have been proposed to address this issue. However, these methods usually obtain the band subset in the perspective of a locally optimal solution. To achieve an optimal solution with a global perspective, this paper developed a novel method for hyperspectral band selection via optimal combination strategy (OCS). The main contributions are as follows: (1) a subspace partitioning approach is proposed which can accurately obtain the partitioning points of the subspace. This ensures that similar bands can be divided into the same subspace; (2) two candidate representative bands with a large amount of information and high similarity are chosen from each subspace, which can fully represent all bands in the subspace; and (3) an optimal combination strategy is designed to acquire the optimal band subset, which achieves an optimal solution with a global perspective. The results on four public datasets illustrate that the proposed method achieves satisfactory performance against other methods.

Determination of sub-ps lattice dynamics in FeRh thin films

Understanding the ultrashort time scale structural dynamics of the FeRh metamagnetic phase transition is a key element in developing a complete explanation of the mechanism driving the evolution from an antiferromagnetic to ferromagnetic state. Using an X-ray free electron laser we determine, with sub-ps time resolution, the time evolution of the (–101) lattice diffraction peak following excitation using a 35 fs laser pulse. The dynamics at higher laser fluence indicates the existence of a transient lattice state distinct from the high temperature ferromagnetic phase. By extracting the lattice temperature and comparing it with values obtained in a quasi-static diffraction measurement, we estimate the electron–phonon coupling in FeRh thin films as a function of laser excitation fluence. A model is presented which demonstrates that the transient state is paramagnetic and can be reached by a subset of the phonon bands. A complete description of the FeRh structural dynamics requires consideration of coupling strength variation across the phonon frequencies.

Novel hyperbolic clustering-based band hierarchy (HCBH) for effective unsupervised band selection of hyperspectral images

For dimensionality reduction of HSI, many clustering-based unsupervised band selection (UBS) methods have been proposed due to their superiority of reducing the high redundancy between selected bands. However, most of these methods fail to reflect the data structure of HSI, leading to inconsistent results of band selection. To tackle this particular issue, we have proposed a novel hyperbolic clustering-based band hierarchy (HCBH) to fully represent the underlying spectral structure and obtain a more consistent band selection. With the proposed adaptive hyperbolic clustering, the performance can be effectively improved with the aid of geometrical information. By introducing a cluster-centre based ranking metric, the desired band subset can be naturally obtained during the clustering process. Experimental results on three popularly used datasets have validated the superior performance of the proposed approach, which outperforms a few state-of-the-art (SOTA) UBS approaches.

Airborne Hyperspectral Imagery for Band Selection Using Moth–Flame Metaheuristic Optimization

In this research, we study a new metaheuristic algorithm called Moth–Flame Optimization (MFO) for hyperspectral band selection. With the hundreds of highly correlated narrow spectral bands, the number of training samples required to train a statistical classifier is high. Thus, the problem is to select a subset of bands without compromising the classification accuracy. One of the ways to solve this problem is to model an objective function that measures class separability and utilize it to arrive at a subset of bands. In this research, we studied MFO to select optimal spectral bands for classification. MFO is inspired by the behavior of moths with respect to flames, which is the navigation method of moths in nature called transverse orientation. In MFO, a moth navigates the search space through a process called transverse orientation by keeping a constant angle with the Moon, which is a compelling strategy for traveling long distances in a straight line, considering that the Moon’s distance from the moth is considerably long. Our research tested MFO on three benchmark hyperspectral datasets—Indian Pines, University of Pavia, and Salinas. MFO produced an Overall Accuracy (OA) of 88.98%, 94.85%, and 97.17%, respectively, on the three datasets. Our experimental results indicate that MFO produces better OA and Kappa when compared to state-of-the-art band selection algorithms such as particle swarm optimization, grey wolf, cuckoo search, and genetic algorithms. The analysis results prove that the proposed approach effectively addresses the spectral band selection problem and provides a high classification accuracy.

Band selection for oxygenation estimation with multispectral/hyperspectral imaging.

Multispectral imaging provides valuable information on tissue composition such as hemoglobin oxygen saturation. However, the real-time application of this technique in interventional medicine can be challenging due to the long acquisition times needed for large amounts of hyperspectral data with hundreds of bands. While this challenge can partially be addressed by choosing a discriminative subset of bands, the band selection methods proposed to date are mainly restricted by the availability of often hard to obtain reference measurements. We address this bottleneck with a new approach to band selection that leverages highly accurate Monte Carlo (MC) simulations. We hypothesize that a so chosen small subset of bands can reproduce or even improve upon the results of a quasi continuous spectral measurement. We further investigate whether novel domain adaptation techniques can address the inevitable domain shift stemming from the use of simulations. Initial results based on in silico and in vivo experiments suggest that 10-20 bands are sufficient to closely reproduce results from spectral measurements with 101 bands in the 500-700 nm range. The investigated domain adaptation technique, which only requires unlabeled in vivo measurements, yielded better results than the pure in silico band selection method. Overall, our method could guide development of fast multispectral imaging systems suited for interventional use without relying on complex hardware setups or manually labeled data.

A principal component analysis of polycyclic aromatic hydrocarbon emission in NGC 7023

ABSTRACT We carried out a principal component analysis (PCA) of the fluxes of five polycyclic aromatic hydrocarbon (PAH) bands at 6.2, 7.7, 8.6, 11.0, and 11.2 µm in the reflection nebula NGC 7023 comprising of the photodissociation region (PDR) and a cavity. We find that only two principal components (PCs) are required to explain the majority of the observed variance in PAH fluxes ( 98 per cent). The first PC ( PC1), which is the primary driver of the variance, represents the total PAH emission. The second PC (PC2) is related to the ionization state of PAHs across the nebula. This is consistent with the results of a similar analysis of the PAH emission in NGC 2023. The biplots and the correlations of PCs with the various PAH ratios show that there are two subsets of ionic bands with the 6.2 and 7.7 µm bands forming one subset and the 8.6 and 11.0 µm bands the other. However, the distinction between these subsets is only present in the PDR. We have also carried out a separate PCA analysis of the PAH fluxes, this time only considering variations in the cavity. This shows that in the cavity, PC2 is not related to the charge state of PAHs but possibly to structural molecular changes.

Multicriteria semi-supervised hyperspectral band selection based on evolutionary multitask optimization

Band selection is a direct and effective method to reduce the spectral dimension, which is one of popular topics in hyperspectral remote sensing. Compared with unsupervised band selection methods, semi-supervised methods seek not only informative but also discriminative band subset by using both labeled and unlabeled samples. However, most currently semi-supervised selection methods simply use a unified criterion on both labeled and unlabeled samples for searching optimal bands, which lacks sample pertinence and adds calculation burden. Since different samples possess different numerical characteristics, optimal criterion on these two kinds of samples may be different. Therefore, a method is required, which can concentrate on the characteristics of labeled and unlabeled samples providing different measure criteria to utilize samples more purposefully. In this paper, a multicriteria semi-supervised model is designed for hyperspectral images band selection. The model is established into two specific tasks: One task measures the amount of information and the redundancy contained in the selected bands from unlabeled samples, the other task utilizes the labeled samples to measure the discrimination of the selected bands. To optimize this model, a multitask optimization strategy is designed to merge the bands information and accelerate the speed of searching the promising bands. In addition, the de-duplication genetic operators are designed to fit the characteristics of hyperspectral images. In this way, the proposed multitask band selection method can select bands with high information, high discrimination, and low redundancy from hyperspectral data in an efficient way according to fully exploiting the numerical characteristics of both labeled and unlabeled samples. Experimental results show the superiority of the proposed method, and demonstrate that the proposed model works more efficiently than the comparison band selection methods.

Deep Reinforcement Learning for Semisupervised Hyperspectral Band Selection

Band selection is an important step in efficient processing of hyperspectral images (HSIs), which can be seen as the combination of powerful band search technique and effective evaluation criterion. The existing deep-learning-based methods make the network parameters sparse to search the spectral bands using threshold-based functions or regularization terms. These methods may lead to an intractable optimization problem. Furthermore, these methods need to repeatedly train deep networks for evaluating candidate band subsets. In this article, we formalize hyperspectral band selection as a reinforcement learning (RL) problem. Band search is regarded as a sequential decision-making process, where each state in the search space is a feasible band subset. To evaluate each state, a semisupervised convolutional neural network (CNN), called EvaluateNet, is constructed by adding the intraclass compactness constraint of both limited labeled and sufficient unlabeled samples. A simple stochastic band sampling method is designed to train EvaluateNet, making it possible to efficiently evaluate without any fine-tuning. In RL, new reward functions are defined by taking the EvaluateNet and the penalty of repeated selection into account. Finally, advantage actor–critic algorithms are designed to explore in the state space and select the band subset according to the expected accumulated reward. The experimental results on HSI data sets demonstrate the effectiveness and efficiency of the proposed algorithms for hyperspectral band selection.