Band Selection Algorithm Research Articles

A restrictive polymorphic ant colony algorithm for the optimal band selection of hyperspectral remote sensing images

ABSTRACTWith hundreds of spectral bands, the rise of the issue of dimensionality in the classification of hyperspectral images is usually inevitable. In this paper, a restrictive polymorphic ant colony algorithm (RPACA) based band selection algorithm (RPACA-BS) was proposed to reduce the dimensionality of hyperspectral images. In the proposed algorithm, both local and global searches were conducted considering band similarity. Moreover, the problem of falling into local optima, due to the selection of similar band subsets although travelling different paths, was solved by varying the pheromone matrix between ants moving in opposite directions. The performance of the proposed RPACA-BS algorithm was evaluated using three public datasets (the Indian Pines, Pavia University and Botswana datasets) based on average overall classification accuracy (OA) and CPU processing time. The experimental results showed that average OA of RPACA-BS was up to 89.80%, 94.96% and 92.17% for the Indian Pines, Pavia University and Botswana dataset, respectively, which was higher than that of the benchmarks, including the ant colony algorithm-based band selection algorithm (ACA-BS), polymorphic ant colony algorithm-based band selection algorithm (PACA-BS) and other band selection methods (e.g. the ant lion optimizer-based band selection algorithm). Meanwhile, the time consumed by RPACA-BS and PACA-BS were slightly lower than that of ACA-BS but obviously lower than that of other benchmarks. The proposed RPACA-BS method is thus able to effectively enhance the search abilities and efficiencies of the ACA-BS and PACA-BS algorithms to handle the complex band selection issue for hyperspectral remotely sensed images.

Kernel-OPBS Algorithm: A Nonlinear Feature Selection Method for Hyperspectral Imagery

The orthogonal-projection-based band selection (OPBS) algorithm is one of the newly proposed band selection methods. In this letter, we present a nonlinear version of the OPBS method, which is denoted as the Kernel-OPBS method. The OPBS method selects the desired bands one by one, and in each round of lookup, it chooses the band that has the maximum distance to the hyperplane spanned by the currently selected bands. Extending this algorithm to a feature space associated with the original input space through a certain nonlinear mapping function can provide a nonlinear version of the OPBS algorithm. Although it is basically intractable to compute the mapped bands due to the high dimensionality of the feature space produced by the nonlinear mapping function, the selection criterion of the Kernel-OPBS method is actually related to only the inner products of the mapped bands; thus, the kernel function can be applied and it is unnecessary to define the nonlinear mapping function. Experimental results on different data sets demonstrate that the selected bands obtained by the Kernel-OPBS method can achieve higher pixel classification performances than that by the OPBS method.

Band selection of hyperspectral image by sparse manifold clustering

Band selection of hyperspectral images is an optimal feature selection method, which aims at reducing the computational burden associated with processing the whole data. The significant and informative bands identified by the band selection process lead to efficient, compact representation of the image data and produce a satisfactory performance in the succeeding applications viz. classification, unmixing, target detection and so on. In this study, the authors present an unsupervised manifold clustering approach for band selection, which accounts for different types of scenarios. Unlike other band selection approaches, the authors’ proposed manifold clustering framework identifies the informative bands by utilising the interrelation between the bands and accounts for the multi-manifold structure prevalent in some real images. The proposed band selection framework identifies the optimal number of clusters by cluster validity index, clusters the bands by manifold clustering and select representative bands from each cluster according to graph weight. Their proposed manifold clustering approach is a generic clustering approach, which produces a satisfactory result even when the data contains non-linearity. The information theoretic performance measures, classification and unmixing performance on real image experiments demonstrate the proficiency of their proposed band selection algorithm.

Structure Tensor and Guided Filtering-Based Algorithm for Hyperspectral Anomaly Detection

Anomaly detection is one of the most important applications of hyperspectral imaging technology. It is a challenging task due to the high dimensionality of hyperspectral images (HSIs), redundant information, noisy bands, and the limited capability of utilizing spatial information. In this paper, we address these problems and propose a novel anomaly detection method in HSIs. Our approach, called structure tensor and guided filter (STGF)-based strategy for anomaly detection, is based on the characteristics of HSIs. First, a novel band selection algorithm is proposed to reduce dimension, remove noisy bands, and select bands with effective information. Second, the selected bands are decomposed into two parts according to the characteristics of anomalies that are usually in a small area. Followed by this step, the backgrounds are removed through a simple differential operation for each selected band. Considering that not all of the bands provide the same contributions to anomaly detection, we then fuse the differential maps by a novel adaptive weighting method to obtain an initial detection map. Finally, GF is conducted to rectify the previous map under the condition that the neighboring pixels usually have quite strong correlations with each other. Experiments have been conducted on real-scene remote sensing HSI. Comparative analyses validate that the proposed STGF method presents superior performance in terms of detection accuracy and computational time.

Empirical study of neighbourhood rough sets based band selection techniques for classification of hyperspectral images

Hyperspectral images are acquired in hundreds of spectral channels that contain rich spectral information of different land-cover objects but at the same time they have high dimensionality with ample redundancy which rises curse of dimensionality and computational issues for classification. To mitigate such issues, band selection that reduces the dimensionality of hyperspectral data is a well-known approach widely used in the literature. Neighbourhood rough set, a variant of rough set capable of analysing continuous values, is a robust mathematical tool for handling uncertain and vague data. In this study, the authors have presented an empirical study of four forward greedy hyperspectral band selection algorithms implemented using the neighbourhood rough set, the variable precision neighbourhood rough set, the consistency measure of neighbourhood rough set and the granulation knowledge-based neighbourhood rough set. The effectiveness of these techniques is compared in terms of average classification accuracy, kappa accuracy and standard deviation obtained by using support vector machine classifier on three real hyperspectral data sets. From the experiments, it is found that the variable precision neighbourhood rough set and the consistency measure of neighbourhood rough set are more robust for selecting informative bands compared to the others. The effectiveness of these techniques is also validated by comparing with some state-of-the-art techniques.

A Coarse-to-Fine Optimization for Hyperspectral Band Selection

Hyperspectral band selection is a feature selection method that selects a most representative set of bands to achieve a good performance in several tasks such as classification and anomaly detection. It reduces the burden of storage, transmission, and computation. In this letter, a two-stage band selection algorithm is introduced. It selects bands and refines the result using a linear reconstruction error criterion. Then a coarse-to-fine band selection (CFBS) strategy is applied to the two-stage band selection in order to achieve a better result. CFBS selects bands group by group. Each group is selected based on bands that are not well represented by the previous groups, trying to minimize the linear reconstruction error. Experiments show that the proposed method has a significant advancement compared with other competitors.

Early Detection of Tomato Spotted Wilt Virus by Hyperspectral Imaging and Outlier Removal Auxiliary Classifier Generative Adversarial Nets (OR-AC-GAN)

Tomato spotted wilt virus is a wide-spread plant disease in the world. It can threaten thousands of plants with a persistent and propagative manner. Early disease detection is expected to be able to control the disease spread, to facilitate management practice, and further to guarantee accompanying economic benefits. Hyperspectral imaging, a powerful remote sensing tool, has been widely applied in different science fields, especially in plant science domain. Rich spectral information makes disease detection possible before visible disease symptoms showing up. In the paper, a new hyperspectral analysis proximal sensing method based on generative adversarial nets (GAN) is proposed, named as outlier removal auxiliary classifier generative adversarial nets (OR-AC-GAN). It is an all-in-one method, which integrates the tasks of plant segmentation, spectrum classification and image classification. The model focuses on image pixels, which can effectively visualize potential plant disease positions, and keep experts’ attention on these diseased pixels. Meanwhile, this new model can improve the performances of classic spectrum band selection methods, including the maximum variance principle component analysis (MVPCA), fast density-peak-based clustering, and similarity-based unsupervised band selection. Selecting spectrum wavebands reasonably is an important preprocessing step in spectroscopy/hyperspectral analysis applications, which can reduce the computation time for potential in-field applications, affect the prediction results and make the hyperspectral analysis results explainable. In the experiment, the hyperspectral reflectance imaging system covers the spectral range from 395 nm to 1005 nm. The proprosed model makes use of 83 bands to do the analysis. The plant level classification accuracy gets 96.25% before visible symptoms shows up. The pixel prediction false positive rate in healthy plants gets as low as 1.47%. Combining the OR-AC-GAN with three existing band selection algorithms, the performance of these band selection models can be significantly improved. Among them, MVPCA can leverage only 8 spectrum bands to get the same plant level classification accuracy as OR-AC-GAN, and the pixel prediction false positive rate in healthy plants is 1.57%, which is also comparable to OR-AC-GAN. This new model can be potentially transferred to other plant diseases detection applications. Its property to boost the performance of existing band selection methods can also accelerate the in-field applications of hyperspectral imaging technology.

Impact of class noise on performance of hyperspectral band selection based on neighborhood rough set theory

Hyperspectral imaging has the ability for characteristic identification because of its rich physical and chemical information. To develop online or portable devices for industrial applications, informative spectral bands relevant to the specific objects are needed to be selected. The neighborhood rough set (NRS) theory is an effective tool for selecting bands. The hyperspectral datasets gathered from practical applications often contain noise. The performance of band selection algorithm is adversely influenced by noise. In this paper, the performance of the band selection algorithm in the presence of class noise was assessed, which based on consistency measure, dependency measure and information measure of the NRS theory. The robustness and classification performance of algorithms for soybean and maize classification using hyperspectral imaging were compared systematically under different levels of noise. The results demonstrate that the robustness and classifying ability of all algorithms decreases in general when the number of mislabeled samples increases in training dataset and testing datasets. Compared with other algorithms based on the NRS theory and some classic algorithms (uninformative variable elimination and successive projections algorithm), the variable precision neighborhood rough set algorithm is proven to be superior. It is more robust and more accurate when class noise occurs in training dataset or testing datasets. This research not only highlights the strong and weak characteristics of the different approaches for different levels of noise, but also provides effective solutions in handling class noise.

MIMR-DGSA: Unsupervised hyperspectral band selection based on information theory and a modified discrete gravitational search algorithm

Band selection plays an important role in hyperspectral data analysis as it can improve the performance of data analysis without losing information about the constitution of the underlying data. We propose a MIMR-DGSA algorithm for band selection by following the Maximum-Information-Minimum-Redundancy (MIMR) criterion that maximises the information carried by individual features of a subset and minimises redundant information between them. Subsets are generated with a modified Discrete Gravitational Search Algorithm (DGSA) where we definine a neighbourhood concept for feature subsets. A fast algorithm for pairwise mutual information calculation that incorporates variable bandwidths of hyperspectral bands called VarBWFastMI is also developed. Classification results on three hyperspectral remote sensing datasets show that the proposed MIMR-DGSA performs similar to the original MIMR with Clonal Selection Algorithm (CSA) but is computationally more efficient and easier to handle as it has fewer parameters for tuning.

Dimensionality reduction based on determinantal point process and singular spectrum analysis for hyperspectral images

Dimensionality reduction is of high importance in hyperspectral data processing, which can effectively reduce the data redundancy and computation time for improved classification accuracy. Band selection and feature extraction methods are two widely used dimensionality reduction techniques. By integrating the advantages of the band selection and feature extraction, the authors propose a new method for reducing the dimension of hyperspectral image data. First, a new and fast band selection algorithm is proposed for hyperspectral images based on an improved determinantal point process (DPP). To reduce the amount of calculation, the dual-DPP is used for fast sampling representative pixels, followed by k-nearest neighbour-based local processing to explore more spatial information. These representative pixel points are used to construct multiple adjacency matrices to describe the correlation between bands based on mutual information. To further improve the classification accuracy, two-dimensional singular spectrum analysis is used for feature extraction from the selected bands. Experiments show that the proposed method can select a low-redundancy and representative band subset, where both data dimension and computation time can be reduced. Furthermore, it also shows that the proposed dimensionality reduction algorithm outperforms a number of state-of-the-art methods in terms of classification accuracy.

Probabilistic rough-set-based band selection method for hyperspectral data classification

This paper proposes an innovative band selection algorithm called probabilistic rough set-based band selection (PRSBS) algorithm. The proposed PRSBS is a supervised band selection algorithm with efficiency for it only needs to calculate the first-order significance measure. The main novelty of the proposed PRSBS algorithm lies in the criterion function which measures the effectiveness of considered band. The PRSBS algorithm uses a probabilistic distribution dependency as the relevance measure between the bands and class labels, which can effectively measure the uncertainty in both the positive and the boundary samples in a dataset. We compared the proposed PRSBS with the most relevant band selection algorithm RSBS on three different hyperspectral datasets, the experimental results show that the PRSBS has better results than the RSBS. Moreover, the PRSBS algorithm runs significantly faster than the RSBS algorithm, which makes it a proper choice for band selection in hyperspectral image dataset.

A Local Potential-Based Clustering Algorithm for Unsupervised Hyperspectral Band Selection

Unsupervised band selection plays an increasingly important role in a hyperspectral image (HSI) classification because of inadequate labeling samples. However, how to select more representative, less redundant, and informative band is an open problem. Recently, a fast density-peak-based clustering (FDPC) algorithm has been proposed. The FDPC chooses the cluster center through the local density and the intracluster distance of each point and ranks each point through a certain rule. For HSI band selection, the FDPC has the following problems. First, when calculating the local density of bands, the difference between bands is not considered, so the local density cannot better characterize the band distribution. Second, the ranking rule in FDPC is more inclined to select bands with larger density and intracluster distance values. However, some boundary bands with abundant information and low redundancy cannot be found. This paper proposes a local potential-based clustering algorithm for unsupervised hyperspectral band selection (LPC). The LPC algorithm improves the FDPC algorithm in three aspects of band selection. First, the local potential of each band is calculated according to the similarity of between bands, and the larger similarity has a greater effect on the local potential. The local potential can better characterize the band distribution. Second, the LPC proposes a weighted ranking rule, which integrates the three factors of local potential, intracluster distance and standard deviation to guide the algorithm to select more discriminative bands. In particular, some boundary bands with abundant information and low redundancy can be found. Finally, an effective method is designed to automatically select the appropriate number of bands. The experimental results on three real hyperspectral data sets demonstrate that the proposed method outperforms the other state-of-the-art methods.

Hyperspectral band selection for soybean classification based on information measure in FRS theory

Soybeans and soy foods have attracted widespread attention due to their health benefits. Special varieties of soybeans are in demand from soybean processing enterprises. Because of the advantage of rapid measurement with minimal sample preparation, hyperspectral imaging technology is used for classifying soybean varieties. Based on fuzzy rough set (FRS) theory, the research of hyperspectral band selection can provide the foundation for variety classification. The performance of band selection with Gaussian membership functions and triangular membership functions under various parameters were explored. Appropriate ranges of parameters were determined by the number of bands and mutual information of subsets relative to the decision. The effectiveness of the proposed algorithms was validated through experiments on soybean hyperspectral datasets by building two classification methods, namely Extreme Learning Machine and Random Forest. Compared with ranking methods, the proposed algorithm provides a promising improvement in classification accuracy by selecting highly informative bands. To further reduce the size of the subset, a post-pruning design was used. For the Gaussian membership function, a subset containing eight bands achieved an average accuracy of 99.11% after post-pruning. As well as classification accuracy, we explored stability of band selection algorithm under small perturbations. The band selection algorithm of the Gaussian membership function was more stable than that of the triangular membership function. The results showed that the information measure (IM) based band selection algorithm is effective at obtaining satisfactory classification accuracy and providing stable results under perturbations.

Band Dual Density Discrimination Analysis for Hyperspectral Image Classification

A novel band discrimination analysis framework for hyperspectral image (HSI) supervised classification is proposed based on dual density (DD). Different from the popular supervised band selection (BS) approaches which measure the discrimination among classes under multivariate normal distribution hypothesis, our work infers the class discrimination degree (overlapping extent) for valid extraction of band subset without any assumed distribution. In the proposed framework, it is crucial to find indexes to measure the discrimination degree of each band, and therefore we develop the DD indexes, including the homogeneity density and the heterogeneity density. Viewing each band of the HSI as a data set, i.e., the data points in each data set are 1-D, and we first obtain the DD value pairs for all data points in each data set. Then, for each data set, we determine its discrimination degree using DD-based zone ratio or score quantify strategy. Finally, the bands, which are determined as the nonoverlapped or have high scores, are chosen as the band subset for the subsequent classification. Superiorities of the proposed BS are demonstrated on the three real-world HSIs over several well-known BS algorithms in terms of classification accuracy and speed.

Enhanced spectrum aggregation based frequency‐band selection routing protocol for cognitive radio ad‐hoc networks

SummaryProviding mobility along with maximum performance in shared spectrum environment is the main objective of Cognitive Radio Systems. Adopting substantial list of parameters to a spectrum aggregation based dynamic route frequency‐band switching is important in Cognitive Radio Mobile Ad‐Hoc Networks (CR‐MANET). Frequent frequency‐band switching affects the performance of the CR‐MANET. In this work geographical position (GPS) detail and spectrum source blind‐spot details are added to the band selection procedure of the Cognitive Radio system. In addition a dedicated Path Prediction Procedure is added with the band selection algorithm to avoid undesirable frequency‐band switching of a mobile CR device. Optimized frequency‐band switching reduces unwanted handovers, thus switching delay and End‐to‐End delay are reduced. Power used to switching between bands is kept in control by sustaining in stable frequency band for longer durations. The combination of GPS data and Path Prediction Procedure is designed to improve the Quality of Service (QoS) parameters like throughput, bandwidth utilization, and Energy efficiency.

A dynamic local cluster ratio-based band selection algorithm for hyperspectral images

Hyperspectral band selection algorithms can save the computational costs during image restoration and analysis. This paper proposes a novel unsupervised band selection method, based on dynamic local cluster ratio (DLCR). The contributions of this paper can be summarized as follows. First, the similarity matrix is calculated in a novel way. Conventional approaches compute the matrix from the Euclidean distances between bands and are vulnerable to noise. Our proposed method can improve the robustness to such noise. Second, we propose an enhanced clustering strategy which clusters each band individually. Third, a dynamic ranking strategy is used to select bands iteratively. Bands that are highly correlated with each other will be prevented from being added to avoid redundancy. DLCR demonstrates improved performance on the Indian Pines and Pavia University data sets, when compared against other methods from the literature.

Automatic band selection algorithm for envelope analysis

Envelope analysis has been widely used to detect early stage faults of rolling element bearings. The primary initial step of envelope analysis is the proper selection of the resonance band for demodulation. Current band selection methods, such as wide band selection, “power spectral density” comparison, and selecting the accelerometer resonance band have limitations such as disturbance of the wide band, the need for a healthy signal for comparison, and the implementation of specialized sensors. In this study, an enhanced method of resonance band selection for envelope analysis was developed. The developed method implements high-pass filtering and “time synchronous averaging” to remove dominant speed-dependent (nonsynchronous and synchronous) spectral contents of a vibration signal. Wavelet packet transform and “root mean square” were then applied to determine the energy distribution of the residual signal. The band with the highest energy (resonance band) was selected for envelope analysis. An experimental study was designed for cross-validation of the developed method. The developed method in this study is more practical than current band selection methods and has no special requirement for sensors. The developed algorithm can be implemented as a processing algorithm in a commercial vibration analyzer, which enhances its ability in early-stage bearing fault detection.

Fast hyperspectral band selection based on spatial feature extraction

Hyperspectral images usually consist of hundreds of spectral bands, which can be used to precisely characterize different land cover types. However, the high dimensionality also has some disadvantages, such as the Hughes effect and a high storage demand. Band selection is an effective method to address these issues. However, most band selection algorithms are conducted with the high-dimensional band images, which will bring high computation complexity and may deteriorate the selection performance. In this paper, spatial feature extraction is used to reduce the dimensionality of band images and improve the band selection performance. The experiment results obtained on three real hyperspectral datasets confirmed that the spatial feature extraction-based approach exhibits more robust classification accuracy when compared with other methods. Besides, the proposed method can dramatically reduce the dimensionality of each band image, which makes it possible for band selection to be implemented in real time situations.

A semi-supervised spatially aware wrapper method for hyperspectral band selection

ABSTRACTBand selection is widely used to identify relevant bands for land-cover classification of hyperspectral images. The combination of spectral and spatial information can improve the classification performance of hyperspectral images dramatically. Similarly, the fusion of spectral–spatial information should also improve the performance of band selection. In this article, two semi-supervised wrapper-based spectral–spatial band selection algorithms are proposed. The local spatial smoothness of hyperspectral imagery is used to improve the performance of band selection when limited labelled samples available. With superpixel segmentation, the first algorithm uses the statistical characteristics of classification map to predict the classification quality of all samples. Based on the Markov random field model, the second algorithm incorporates the spatial information by the minimization of spectral–spatial energy function. Four widely used real hyperspectral data sets are used to demonstrate the effectiveness of the proposed methods, when compared to cross-validation-based wrapper method, the accuracy is improved by 2% for different data sets.

Progressive Sample Processing of Band Selection for Hyperspectral Image Transmission

Band selection (BS) is one of the important topics in hyperspectral image (HSI) processing. Many types of BS algorithms were proposed in the last decade. However, most of them were designed for off-line use. They can only be used with pre-collected data, and are sometimes ineffective for applications that require timeliness, such as disaster prevention or target detection. This paper proposes an online BS method that allows us obtain instant BS results in a progressive manner during HSI data transmission, which is carried out under band-interleaved-by-sample/pixel (BIS/BIP) format. Such a revolutionary method is called progressive sample processing of band selection (PSP-BS). In PSP-BS, BS can be done recursively pixel by pixel, so that the instantaneous BS can be achieved without waiting for all the pixels of an image. To develop a PSP-BS algorithm, we proposed PSP-OMPBS, which adopted the recursive version of a self-sparse regression BS method (OMPBS) as a native algorithm. The experiments conducted on two real hyperspectral images demonstrate that PSP-OMPBS can progressively output the BS with extremely low computing time. In addition, the convergence of BS results during transmission can be further accelerated by using a pre-defined pixel transmission sequence. Such a significant advantage not only allows BS to be done in a real-time manner for the future satellite data downlink, but also determines the BS results in advance, without waiting to receive every pixel of an image.