Search

Band selection is crucial for achieving high classification accuracy in hyperspectral image (HSI) analysis, especially when ground truth data are limited. While unsupervised algorithms are preferred in such scenarios, the effectiveness of k-means clustering depends heavily on the choice of similarity measure. This article presents a novel two-level clustering approach for band selection. In the first level, bands are clustered using k-means with various similarity measures such as Euclidean distance, spectral angle mapper (SAM), and spectral information divergence (SID). Subsequently, the second level leverages the chord metric k-means clustering to form clusters of HSI scenes upon optimal band clusters from the first level. This initial band selection reduces dimensionality and guides subsequent k-means clustering. The proposed chord-based clustering method, utilizing the chord metric, outperforms standard k-means variants, demonstrating significant improvements in accuracy. Experimental results on publicly available hyperspectral datasets confirm the effectiveness of the proposed approach as an alternative to traditional k-means algorithms, showcasing significant improvements in accuracy. [ABSTRACT FROM AUTHOR]

Rapid advancement in satellite communication over the last decade have resulted in the widespread use of remote sensing images. Additionally, as satellite image transmission over the Internet has increased, secrecy concerns have also arisen. As a result, digitally transmitted images must have great imperceptibility and confidentiality. Multispectral images consist of multiple bands. It is very challenging to select the important spectral band for watermarking so that the structural and visual quality of the satellite Image can be retained. This work proposes an innovative blind watermarking model based on a hybrid optimization strategy performed with the following two processes: the embedding process and the extraction process. A novel hybrid optimization named FBIAO algorithm, which is the amalgamation of Archimedes Optimization (ArchOA) and Forensic Based Investigation Optimization (FBIO) algorithm is used to select spectral band for watermarking. The proposed novel FBIAO enhances the balances between the exploration and exploitation, boosts the solution diversity and improves the convergence of FBI based optimization for spectral band selection. The 3-level Discrete Wavelet Transform (DWT) is used to embed the watermark logo in the selected spectral band image and then position selection is applied to identify the location for embedding the watermark. Further, the watermark image is scrambled using Arnold Map technique to avoid the correlation between image pixel. The proposed method provides a peak signal-to-noise ratio (PSNR) in the range of 35.57 dB to 36.80 dB and, a structural similarity index (SSIM) between 0.91 to 0.93 without attack for six sample datasets. It provides robustness for different attacks and offers SSIM in between 0.6 to 0.87 and normalized Correlation (NC) in between 0.8 to 0.91 which is superior over traditional techniques. [ABSTRACT FROM AUTHOR]

Hyperspectral imaging has appeared as a vital tool in remote sensing science for its efficacy in effectively delineating regions of interest. However, the classification of hyperspectral images (HSI) encounters notable challenges, including the high dimensionality of highly correlated bands and the scarcity of training samples. Addressing these challenges is very essential by determining the most relevant bands, as well as the utilization of unlabelled training samples. In response to these issues, this study presents an unsupervised framework based on an enhanced Mayfly Optimization Algorithm (MOA) in order to select the most informative spectral bands. The enhanced MOA effectively identifies informative bands by leveraging the random solutions to explore the global search space, and enhance the solution diversity. On the other hand, leveraging the best experiences to boost the local search, efficiently attaining optimal solutions. This balanced exploration-exploitation strategy ensures the algorithm's robustness and effectiveness in addressing the optimization problem. Ultimately, the proposed approach is demonstrated at the pixel-level hyperspectral image classification using two machine learning classifiers: Random Forest and Support Vector Machine. Thorough experimentation carried out on three benchmark hyperspectral datasets consistently confirms the effectiveness of the proposed approach. [ABSTRACT FROM AUTHOR]

Soil water content (SWC) is an important reference for reflecting agricultural drought conditions and irrigation management. Accurate acquisition and monitoring of SWC information is crucial for water-saving agriculture and soil protection. Optical remote sensing technology has the advantages of high efficiency, continuity, noncontact, and wide regional coverage, which is one of the most commonly used SWC estimating methods. First, this work studies the best modeling band and puts forward the band that is sensitive to SWC but insensitive to the illumination and view angle as the modeling band. Then, the process of the internal reflection of incident radiation caused by the water layer around soil particles is added to improve the SWAP-Hapke model. This work innovatively proposed using absorbance instead of reflectance to select the modeling band and establish an improved SWAP-Hapke model. In addition, a complete multiangle and multi-SWC gradient dataset is used to improve the estimation accuracy of model parameters. Finally, the multiangle retrieval accuracy of the improved SWAP-Hapke model was evaluated. The results show that the improved SWAP-Hapke model based on absorbance has obviously higher retrieval accuracy in different view and illumination directions, and its accuracy is less affected by the illumination and view geometry. The average R2 of the improved SWAP-Hapke model established by using reflectance and absorbance can reach 0.89 and 0.92, and the average of the root-mean-square error is 4% and 2.4%. In addition, the five optimal bands for establishing the SWC retrieval model are the absorbance at 1989, 1462, 2357, 2190 nm, and reflectance at 1460 nm. Finally, the improved SWAP-Hapke model is applied to the public AAV dataset, and the retrieval results are relatively consistent with measured SWC.

Recently, graph learning-based hyperspectral band selection algorithms illustrate impressive performance for hyperspectral image (HSI) processing, whose goal is to select an optimal band combination containing less redundancy through the learned graph matrix. However, most of the previous methods work in single spectral domain, which neglects the rich image spatial information. Moreover, they typically model the graph matrix from a global perspective while ignoring the differences in spatial distribution that exist in diverse pixel and band regions. Based on the above considerations, to take full account of structure information in spatial and spectral domains, a structural graph learning method for hyperspectral band selection (SGLM) is designed. Specifically, SGLM constructs two matrices using the correlation between pixels and bands under the local perspective, which can capture the image structure information reasonably. Since the proposed method is modelled in both spatial and spectral dimensions, it is conducive to subsequent band subset generation task. Meanwhile, in order to guarantee local spatial consistency among bands, a Laplacian regularization term associated with the error matrix is introduced to the self-representation model. Additionally, considering that the importance of a band is consistent with its capability to reconstruct the entire band set, an adjacent bands reconstruction strategy is adopted to obtain the ultimate band combination, which can assess the significance of each band effectively. The comprehensive experimental analysis on four datasets demonstrates that the proposed SGLM model has outstanding superiority in comparison with several band selection algorithms. [ABSTRACT FROM AUTHOR]

Along with the high spectral rich information it provides, one of the difficulties in processing a hyperspectral image is the need for expert knowledge and high-spec hardware to process very high-dimensional data. 3D convolutional neural network (3D CNN), which uses spectral and spatial features together, enables a powerful solution for HSI classification. This study proposes an efficient dual-stream 3D CNN for accurate HSI classification. The proposed method offers effective classification using spectral-spatial features without relying on pre-processing or post-processing. A comparative study of how CNN classification performance is affected by hyperspectral band selection based on deep reinforcement learning (DRL) is presented. Using the most relevant bands in the hyperspectral image is decisive in deep CNN networks without losing information and accuracy. The proposed method was compared with 3D CNN, 3D + 1D CNN, Multiscale 3D deep convolutional neural network (M3D-DCNN), and InceptionV3 algorithms using Indian Pines (IP), Salinas, Pavia Center (PaviaC), Houston 2013 and QUH-Tangdaowan datasets. It achieved 92.43 % overall accuracy (OA) in IP, 95.06 % OA in Salinas dataset, 99.00 % OA in PaviaC dataset, 91.25 % OA in Houston 2013 and 94.87 % OA in QUH-Tangdaowan. Codes are released at: https://github.com/lapistlazuli/DS-3DCNN. [ABSTRACT FROM AUTHOR]

In environmental studies, rhodamine dyes are commonly used to trace water movements and pollutant dispersion. Remote sensing techniques offer a promising approach to detecting rhodamine and estimating its concentration, enhancing our understanding of water dynamics. However, research is needed to address more complex environments, particularly optically shallow waters, where bottom reflectance can significantly influence the spectral response of the rhodamine. Therefore, this study proposes a novel approach: transferring pre-trained classifiers to develop a generalizable method across different environmental conditions without the need for in situ calibration. Various samples incorporating distilled and seawater on light and dark backgrounds were analyzed. Spectral analysis identified critical detection regions (400–500 nm and 550–650 nm) for estimating rhodamine concentration. Significant spectral variations were observed between light and dark backgrounds, highlighting the necessity for precise background characterization in shallow waters. Enhanced by the Sequential Feature Selector, classification models achieved robust accuracy (>90%) in distinguishing rhodamine concentrations, particularly effective under controlled laboratory conditions. While band transfer was successful (>80%), the transfer of pre-trained models posed a challenge. Strategies such as combining diverse sample sets and applying the first derivative prevent overfitting and improved model generalizability, surpassing 85% accuracy across three of the four scenarios. Therefore, the methodology provides us with a generalizable classifier that can be used across various scenarios without requiring recalibration. Future research aims to expand dataset variability and enhance model applicability across diverse environmental conditions, thereby advancing remote sensing capabilities in water dynamics, environmental monitoring and pollution control. [ABSTRACT FROM AUTHOR]

Band selection (BS) aims to reduce redundancy in hyperspectral imagery (HSI). Existing BS approaches typically model HSI only in a single dimension, either spectral or spatial, without exploring the interactions between different dimensions. To this end, we propose an unsupervised BS method based on a spectral–spatial cross-dimensional attention network, named SSANet-BS. This network is comprised of three stages: a band attention module (BAM) that employs an attention mechanism to adaptively identify and select highly significant bands; two parallel spectral–spatial attention modules (SSAMs), which fuse complex spectral–spatial structural information across dimensions in HSI; a multi-scale reconstruction network that learns spectral–spatial nonlinear dependencies in the SSAM-fusion image at various scales and guides the BAM weights to automatically converge to the target bands via backpropagation. The three-stage structure of SSANet-BS enables the BAM weights to fully represent the saliency of the bands, thereby valuable bands are obtained automatically. Experimental results on four real hyperspectral datasets demonstrate the effectiveness of SSANet-BS. [ABSTRACT FROM AUTHOR]

Skin cancer is the predominant form of cancer worldwide, including 75% of all cancer cases. This study aims to evaluate the effectiveness of the spectrum-aided visual enhancer (SAVE) in detecting skin cancer. This paper presents the development of a novel algorithm for snapshot hyperspectral conversion, capable of converting RGB images into hyperspectral images (HSI). The integration of band selection with HSI has facilitated the identification of a set of narrow band images (NBI) from the RGB images. This study utilizes various iterations of the You Only Look Once (YOLO) machine learning (ML) framework to assess the precision, recall, and mean average precision in the detection of skin cancer. YOLO is commonly preferred in medical diagnostics due to its real-time processing speed and accuracy, which are essential for delivering effective and efficient patient care. The precision, recall, and mean average precision (mAP) of the SAVE images show a notable enhancement in comparison to the RGB images. This work has the potential to greatly enhance the efficiency of skin cancer detection, as well as improve early detection rates and diagnostic accuracy. Consequently, it may lead to a reduction in both morbidity and mortality rates. [ABSTRACT FROM AUTHOR]

Hyperspectral imaging (HSI) is essential for capturing images across various spectral bands and providing indepth spectral information. However, the high dimensionality of HSI data presents challenges, such as increased computational complexity and the curse of dimensionality. Band selection is a critical pre-processing step that addresses these challenges by identifying the most informative bands. This paper introduces a novel band selection method utilizing the Tabu Search algorithm (TSA), aimed at optimizing the selection of spectral bands to enhance classification performance. The proposed method assesses bands using a fitness function that maximizes variance and minimizes correlation among the chosen bands. Experimental results on Indian pine dataset and Pavia University Scenes, along with the K-Nearest Neighbor and Support Vector machine improves the classification accuracy while reducing computational demands. [ABSTRACT FROM AUTHOR]

Precise spatial mapping of individual species using hyperspectral data is crucial for effective forest management and policy-making. This study focuses on Rhododendron arboreum, known for its medicinal properties attributed to the flavonoid Quercitrin. Sample data and spectroradiometer data were collected from the complex terrain of the Kumaon region in the Himalayas. Hyperspectral data, which includes signal variations based on biophysical and biochemical properties along with noise, were preprocessed using filtering techniques to enhance signal clarity by removing noise. Smoothing techniques were applied to remove noisy bands from the spectra, such as the Savitzky-Golay filter for reduced least square fit complexity and the Average Mean filter for taking mean spectral values. Subsequently, Spectral Analysis (SA) techniques, including first derivative, second derivative, and continuum removal, were employed. These mathematical transformations highlighted absorption troughs and determined the effect of Quercitrin on spectral wavelengths. Principal Component Analysis (PCA) was used to identify the most relevant bands related to Quercitrin. Additionally, regression analysis was applied on resampled spectral data, selected significant wavelengths based on variable importance values, pinpointing the most prominent wavelengths: 1196, 1229, 1328, 1383, 1425, 1636, 1661, 1699, 1785, and 1715 nm. Over 50 two-band combination indices were tested, and those with p-values less than 0.05 were deemed significant. For the development of prediction model, Machine Learning (ML) algorithms, including Support Vector Machine (SVM), Relevance Vector Machine (RVM), Random Forest (RF), and Artificial Neural Network (ANN), were applied. The Random Forest model, which splits input data into trees to simulate the best model based on observed values, demonstrated high effectiveness in predicting Quercitrin levels, achieving a training correlation of 0.864 and a testing correlation of 0.570. Hence RF proved to be a best technique of band selection as well as robust for Quercitrin prediction. This methodological approach highlights the importance of advanced data processing and analysis techniques in remote sensing applications for forest phytochemical prediction.

Abstract As one of the dimensionality reduction techniques of hyperspectral image (HSI), band selection (BS) does not change the spectral characteristics and physical meaning of HSIs, which is beneficial to the identification and analysis of surface objects. Recently, many BS methods for target detection have achieved promising results by making full use of the priori spectral features of the target to be detected. Conversely, anomaly detection separates the anomaly based solely on the statistical distribution difference between anomaly and background without any prior information. Therefore, the development of BS for anomaly detection has lagged far behind that of BS for target detection. To this end, this paper proposes a novel BS algorithm dedicated to anomaly detection tasks, named anomaly‐background separation and particle swarm optimization (PSO)‐based BS. Specifically, an anomaly‐background separation framework (ABSF) is established to predetermine a priori knowledge of anomaly distribution. Then, three band prioritization criteria are constructed with the anomaly‐background constraints generated by ABSF. Finally, PSO is used to find the optimal subset of bands in the solution space. The experiments on two real datasets demonstrate that the proposed method yields better detection results and greater stability compared to other BS methods discussed in this paper.

With the rapid development of information technology, the traditional information management system faces the challenge of efficiency and accuracy when dealing with complex data and decision-making problems. In order to solve these problems, this paper integrates the random leapfrog band selection algorithm into the information management system. This algorithm shows excellent performance in dealing with large-scale, nonlinear and multi-variable optimization problems, resource allocation, scheduling optimization and data mining. This paper mainly explains the design framework of information management system, the optimization of information management by random leap-band selection algorithm and the design of information management system. Finally, two groups of simulation experiment results are as follows: the accuracy of the optimized information management system is 96%, the running time is 7.1min, the algorithm performance is better than the traditional system, and the response time of the system is reduced by 195ms. This research provides an efficient, flexible and scalable solution for dealing with various complex information management system optimization problems. [ABSTRACT FROM AUTHOR]

As one of the dimensionality reduction techniques of hyperspectral image (HSI), band selection (BS) does not change the spectral characteristics and physical meaning of HSIs, which is beneficial to the identification and analysis of surface objects. Recently, many BS methods for target detection have achieved promising results by making full use of the priori spectral features of the target to be detected. Conversely, anomaly detection separates the anomaly based solely on the statistical distribution difference between anomaly and background without any prior information. Therefore, the development of BS for anomaly detection has lagged far behind that of BS for target detection. To this end, this paper proposes a novel BS algorithm dedicated to anomaly detection tasks, named anomaly‐background separation and particle swarm optimization (PSO)‐based BS. Specifically, an anomaly‐background separation framework (ABSF) is established to predetermine a priori knowledge of anomaly distribution. Then, three band prioritization criteria are constructed with the anomaly‐background constraints generated by ABSF. Finally, PSO is used to find the optimal subset of bands in the solution space. The experiments on two real datasets demonstrate that the proposed method yields better detection results and greater stability compared to other BS methods discussed in this paper. [ABSTRACT FROM AUTHOR]

To develop rice varieties with better nutritional qualities, it is important to classify rice seeds accurately. Hyperspectral imaging can be used to extract spectral information from rice seeds, which can then be used to classify them into different varieties. The challenges of precise classification increase when there are many classes and few training samples. In this paper, we present a novel method for high-precision Hyperspectral Image (HSI) classification of 90 different classes of rice seeds using ensemble deep learning. Our method first employs band selection techniques to select the optimal hyperspectral bands for rice seed classification. Then, a deep neural network is trained with the selected hyperspectral and RGB data from rice seed images to obtain different models for different bands. Finally, an ensemble of deep learning models is employed to classify rice seed images and improve classification accuracy. The proposed method achieves an overall precision ranging from 92.73 to 96.17% despite a large number of classes and low data samples for each class and with only 15 selected hyperspectral bands. This precision is significantly higher than the state-of-the-art classical machine learning methods like random forest, confirming the effectiveness of the proposed method in classifying hyperspectral images of rice seeds. [ABSTRACT FROM AUTHOR]

Hyperspectral (HS) image classification plays a crucial role in numerous areas including remote sensing (RS), agriculture, and the monitoring of the environment. Optimal band selection in HS images is crucial for improving the efficiency and accuracy of image classification. This process involves selecting the most informative spectral bands, which leads to a reduction in data volume. Focusing on these key bands also enhances the accuracy of classification algorithms, as redundant or irrelevant bands, which can introduce noise and lower model performance, are excluded. In this paper, we propose an approach for HS image classification using deep Q learning (DQL) and a novel multi-objective binary grey wolf optimizer (MOBGWO). We investigate the MOBGWO for optimal band selection to further enhance the accuracy of HS image classification. In the suggested MOBGWO, a new sigmoid function is introduced as a transfer function to modify the wolves' position. The primary objective of this classification is to reduce the number of bands while maximizing classification accuracy. To evaluate the effectiveness of our approach, we conducted experiments on publicly available HS image datasets, including Pavia University, Washington Mall, and Indian Pines datasets. We compared the performance of our proposed method with several state-of-the-art deep learning (DL) and machine learning (ML) algorithms, including long short-term memory (LSTM), deep neural network (DNN), recurrent neural network (RNN), support vector machine (SVM), and random forest (RF). Our experimental results demonstrate that the Hybrid MOBGWO-DQL significantly improves classification accuracy compared to traditional optimization and DL techniques. MOBGWO-DQL shows greater accuracy in classifying most categories in both datasets used. For the Indian Pine dataset, the MOBGWODQL architecture achieved a kappa coefficient (KC) of 97.68% and an overall accuracy (OA) of 94.32%. This was accompanied by the lowest root mean square error (RMSE) of 0.94, indicating very precise predictions with minimal error. In the case of the Pavia University dataset, the MOBGWO-DQL model demonstrated outstanding performance with the highest KC of 98.72% and an impressive OA of 96.01%. It also recorded the lowest RMSE at 0.63, reinforcing its accuracy in predictions. The results clearly demonstrate that the proposed MOBGWODQL architecture not only reaches a highly accurate model more quickly but also maintains superior performance throughout the training process. [ABSTRACT FROM AUTHOR]

Hyperspectral imagery encompasses spectral and spatial dimensions, reflecting the material properties of objects. Its application proves crucial in search and rescue, concealed target identification, and crop growth analysis. Clustering is an important method of hyperspectral analysis. The vast data volume of hyperspectral imagery, coupled with redundant information, poses significant challenges in swiftly and accurately extracting features for subsequent analysis. The current hyperspectral feature clustering methods, which are mostly studied from space or spectrum, do not have strong interpretability, resulting in poor comprehensibility of the algorithm. So, this research introduces a feature clustering algorithm for hyperspectral imagery from an interpretability perspective. It commences with a simulated perception process, proposing an interpretable band selection algorithm to reduce data dimensions. Following this, amulti-dimensional clustering algorithm, rooted in fuzzy and kernel clustering, is developed to highlight intra-class similarities and inter-class differences. An optimized P systemis then introduced to enhance computational efficiency. This system coordinates all cells within a mapping space to compute optimal cluster centers, facilitating parallel computation. This approach diminishes sensitivity to initial cluster centers and augments global search capabilities, thus preventing entrapment in local minima and enhancing clustering performance. Experiments conducted on 300 datasets, comprising both real and simulated data. The results show that the average accuracy (ACC) of the proposed algorithm is 0.86 and the combination measure (CM) is 0.81. [ABSTRACT FROM AUTHOR]

The technological advancements in spectroscopy give rise to acquiring data about different materials on earth's surface which can be utilized in a variety of potential applications. But, the hundreds of spectral bands are generally equipped with highly correlated information with limited training samples. This will degrade the Hyperspectral Image (HSI) classification accuracy. So Dimensionality Reduction (DR) has become inevitable and necessary step need to incorporate before HSI classification. The main contribution of this work lies in comparative study and review on dimensionality reduction techniques for Hyperspectral remote sensing image classification. The related challenges and research directions are also discussed. This study will help the researchers in the Hyperspectral remote sensing community to choose the appropriate DR technique for classification which can be useful in various real time applications.

Land Use/Land Cover (LULC) classification using hyperspectral images in remote sensing is a leading technology. However, LULC classification using hyperspectral images is a difficult task and time-consuming process because it has fewer training samples. To overcome these issues, we proposed a deep-Long Short-Term Memory (deep-LSTM) to classify the LULC. Before classifying the LULC, extracting valuable features from an image is needed, and after extracting the features, selecting the bands which are helpful for classification should be done. In this work, we have proposed an auto-encoder model for feature extraction, a ranking-based band selection model to select the bands, and deep-LSTM for classification. We have used three publicly available benchmark datasets; they are Pavia University (PU), Kennedy Space Centre (KSC), and Indian Pines (IP). Average Accuracy (AA), Overall Accuracy (OA), and Kappa Coefficient (KC) are used to measure the classification accuracy. The suggested technique has provided the top outcomes compared to the other state-of-the-art methods.

Band selection is a key strategy to address the challenges of managing large hyperspectral datasets and reduce the dimensionality problem associated with the simultaneous analysis of hundreds of spectral bands. However, the computational complexity of traditional methods makes the algorithms difficult to be deployed on board satellites. This is especially true for small satellites with limited computational and power resources. Moreover, the existing band selection techniques often require the hypercube to be processed at least at Level-1B product, i.e., the bands need to be finely aligned before selecting them, demanding more computational resources for the onboard computer. This study presents a novel neural-network-based approach for onboard band selection using data with coarse band-to-band aligned. This methodology not only simplifies the preprocessing requirements but also opens new possibilities for efficient hyperspectral imaging from space onboard small satellites, such as classification, change, and target detection.

Hyperspectral imaging is a key tool in numerous remote sensing applications. Images with tens or even hundreds of spectral bands contain a wealth of information to retrieve multiple geophysical parameters, to perform target detection, or land classification with unbeatable high accuracies. In addition, hyperspectral sensors are becoming smaller, and today they even fit in CubeSats. However, the amount of data they generate is so large that satellite communication systems have severe limitations to download it, especially in SmallSats. It is therefore becoming urgent to develop efficient automated algorithms that can be executed in the limited capabilities of the onboard computers of these satellites, so as to reduce the amount of data to be stored and downloaded, while keeping as much information as possible for a given scene. In this work, a band selection algorithm has been designed to deal with this problem. The proposed algorithm consists of the sequential selection of the spectral bands ranked using the amount of information provided by each band, and also the correlation of these bands with the previously selected ones. The algorithm performance is assessed by means of a suite of classification tests with hyperspectral datasets from different sensors. Results show comparable or even better performance than other existing band selection algorithms, while outperforming in terms of computational complexity, which makes it more suitable for SmallSats with limited computing resources.

A Hyperspectral image (HSI) contains numerous spectral bands, providing better differentiation of ground objects. Although the data from HSI are very rich in information, their processing presents some difficulties in terms of computational effort and reduction of information redundancy. These difficulties stem mainly from the fact that the HSI consists of a large number of bands along with some redundant bands. Band selection (BS) is used to select a subset of bands to reduce processing costs and eliminate spectral redundancy. BS methods based on a metaheuristic approach have become popular in recent years. However, most BS methods based on a metaheuristic approach can get stuck in the local optimum and converge slowly due to a lack of balance between exploration and exploitation. In this paper, three BS methods are proposed for HSI data. The first method applies Crow Search Algorithm (CSA) for BS. The other two proposed methods, HPSOCSA_SP and HPSOCSA_SLP, are based on the hybridization of Particle Swarm Optimization (PSO) and CSA. The purpose of these hybridizations is to balance exploration and exploitation in a search process for optimal band selection and fast convergence. In hybridization techniques, PSO and CSA exchange informative data at each iteration. HPSOCSA_SP split the population into two equal parts. PSO is applied to one part and CSA to the other. HPSOCSA_SLP selects half of the top-performing members based on fitness. PSO and CSA are applied to the selected population sequentially. Our proposed models underwent rigorous testing on four HSI datasets and showed superiority over other metaheuristic techniques. [ABSTRACT FROM AUTHOR]

Hyperspectral imaging effectively distinguishes land cover classes with discriminative spectral-spatial characteristics. However, hyperspectral image (HSI) classification encounters three significant challenges: the curse of dimensionality, the scarcity of training samples, and the necessity to incorporate spatial information within the classification process. The large number of redundant and irrelevant spectral bands in HSI can negatively impact the classification performance. Furthermore, the annotation process for training samples is expensive and time-consuming. To address these challenges, this work presents a novel semi-supervised approach for hyperspectral image dimensionality reduction using the Mayfly Algorithm (MA). The proposed method aims to preserve the most informative spectral bands while leveraging both labelled and unlabelled samples in the process. The MA algorithm effectively balances exploration and exploitation strategies, enhancing population diversity and mitigating challenges such as premature convergence and local optima. Additionally, the inclusion of structural variation and thresholding (SVT) in the Bitonic filter enables the incorporation of spatial features during the classification phase. This improves the separability of pixels belonging to different objects and leads to significant improvements in classification performance. Finally, two machine learning classifiers, namely Random Forest (RF) and Support Vector Machine (SVM), are applied at the pixel level for hyperspectral image classification. Extensive experimental results on three real hyperspectral datasets demonstrate the superiority of the proposed approach compared to state-of-the-art algorithms like Harmony Search (HS), Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and more with an average improvement of 3.2%. [ABSTRACT FROM AUTHOR]

Forest pests impose significant threats to tree productivity and ecosystem stability. Most studies focussed on foliage water, tree defoliate and mortality, which lag behind pest outbreak. To address this issue, we developed hyperspectral LiDAR (HSL) system based on an acousto-optic tunable filter (AOTF) to examine the echo intensity of tree trunks. First, we demonstrated the echo power of the cylindrical surface is approximately equivalent to that of the plate surface within 60° incidence angle, which provide the theoretical reference for trunk surface echo intensity calibration with a white panel. Second, we propose a two-step method to detect the attacked parts from barks. The first step is rapid pest detection conducted by the echo intensity ratio of damaged area and normal area. While the ratio is around one, the second step is pest detection based on support vector machine classifier with reflectance. When the reflectivity of five spectral channels is used as input parameters, 100% rapid detection accuracy of damage area is achieved. We also use the radiance under seven channels to achieve 100% classification of tree species and attack features. The results show that this HSL is effective in wood-boring pest detection and contributes to forest applications. [ABSTRACT FROM AUTHOR]

There is information redundancy in both spatial and spectral aspects of hyperspectral images. Considering a fixed proportion in sequential forward method may not find the optimal bands, we modify a band selection (BS) method by introducing a parameter to adjust the proportion of standard deviation to correlation, which may select key bands quickly and accurately. In addition, slant Haar type orthogonal transforms (SHTOT) have slant base vectors suitable to express the image brightness with gradual change. However, SHTOT have attracted little attention of scholars'. This paper introduces SHTOT with fast algorithm and varied parameters to the further compression of band images from a space point of view. Comparative experiments were performed with other BS strategies and state-of-the-art orthogonal transforms, such as DCT, DWT, Walsh, slant transform, Haar type orthogonal transforms. Final experimental results achieved on the commonly used data sets validate that the proposed approach has a faster speed, high compression ratio and good image quality. Additionally, it is more appropriate to choose different SHTOT for the specific application. The sparse SHTOT generated by certain parameter values are more suitable for the high requirement in image quality, but the compression ratio isn't very high, while dense SHTOT produced by another parameter values are fitter for the opposite cases. [ABSTRACT FROM AUTHOR]

The technological advancements in spectroscopy give rise to acquiring data about different materials on earth's surface which can be utilized in a variety of potential applications. But, the hundreds of spectral bands are generally equipped with highly correlated information with limited training samples. This will degrade the Hyperspectral Image (HSI) classification accuracy. So Dimensionality Reduction (DR) has become inevitable and necessary step need to incorporate before HSI classification. The main contribution of this work lies in comparative study and review on dimensionality reduction techniques for Hyperspectral remote sensing image classification. The related challenges and research directions are also discussed. This study will help the researchers in the Hyperspectral remote sensing community to choose the appropriate DR technique for classification which can be useful in various real time applications. [ABSTRACT FROM AUTHOR]

Land Use/Land Cover (LULC) classification using hyperspectral images in remote sensing is a leading technology. However, LULC classification using hyperspectral images is a difficult task and time-consuming process because it has fewer training samples. To overcome these issues, we proposed a deep-Long Short-Term Memory (deep-LSTM) to classify the LULC. Before classifying the LULC, extracting valuable features from an image is needed, and after extracting the features, selecting the bands which are helpful for classification should be done. In this work, we have proposed an auto-encoder model for feature extraction, a ranking-based band selection model to select the bands, and deep-LSTM for classification. We have used three publicly available benchmark datasets; they are Pavia University (PU), Kennedy Space Centre (KSC), and Indian Pines (IP). Average Accuracy (AA), Overall Accuracy (OA), and Kappa Coefficient (KC) are used to measure the classification accuracy. The suggested technique has provided the top outcomes compared to the other state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Unsupervised hyperspectral image (HSI) band selection methods have been attracting ever-increasing attention. However, the local structural features captured by most of the existing methods suffer from a certain degree of risk of feature ambiguity. Moreover, these methods typically have difficulty preserving the latent manifold structure of the HSI in selected subsets of the bands. To address these problems, firstly, the graph approximate representation learning (GARL) model is proposed in this article by incorporating graph-regularized sparse coding into self-representation learning to preserve the local manifold structure information of HSI in the selected subset of bands. Meanwhile, to alleviate the ambiguity of the local graph structures, we design a local-global feature relationship expression model to construct the latent long-distance contextual connectivity (LDCC) graphs between the obtained local graph structures. Then, to preserve HSI's local and non-local manifold structure information as much as possible in the selected subset of bands, a multigraph approximate representation learning (MGARL) model is proposed in this article by incorporating the obtained LDCC graph into the proposed GARL model. Next, we design a solution method to solve the proposed formulation. Finally, extensive experimental results on four datasets reveal the promising performance of MGARL over some state-of-the-art methods. [ABSTRACT FROM AUTHOR]

A popular strategy to reduce the spectral size of HyperSpectral Images (HSI) is to suppress all bands whose information content is largely duplicated in surviving bands. This is commonly termed Band Selection (BS). In this manuscript, a novel, unsupervised BS technique is proposed, consisting of three main steps: 1) Adjacent structural-similarity-based band partitioning, 2) for each partition, correlation-based, cluster-representative band selection, and 3) bad cluster removal using an Entropy measure. Initially, the adjacent structural similarity is measured, using Structural SIMilarity Index (SSIM); the next step is adjacency-partitioned cluster formation using a derivative-based approach. Then, a representative band with a high correlation is selected for each cluster. Finally, according to the Entropy measure, the cluster representatives are ranked to obtain the optimal subset of bands. The proposed Adjacency Clustering and Local Structural Correlation (ACLSC)-based BS is compared with other state-of-art approaches in terms of classification performance on benchmark HSI datasets. Overall Accuracy (OA), Average Accuracy (AA), and Kappa ($\kappa $ κ) are computed on three datasets, namely, Indian Pines, Salinas, and Pavia University, to validate the proposed algorithm. Validation results support the use of adjacency clustering and incorporation of spatial-structural features for BS purposes. [ABSTRACT FROM AUTHOR]

Dimensionality reduction is a prevalent preprocessing step when analysing hyperspectral images (HSIs). It uses band selection (BS) and feature extraction (FE) techniques, but the former can retain the physical meaning and interpretability of the HSI datasets. Hence, these techniques are preferred to select a proper subset of bands without transformation. This letter introduces a straightforward supervised BS technique using discriminatory groups (DGs) that does not require complex computations. In this technique, total DGs are achieved equal to the spectral dimension of the HSI dataset, and each holds bands similar to the number of classes. The constituent bands of the DGs may present more than once, but the total number of bands in the group remains unchanged. The obtained DGs are ranked using high- to low-discrimination fractions, which indicate the uniqueness of the respective DG. The highest value that a DG can attain represents the group's ability to distinguish among all classes (the maximum number of classes). Hence, the proposed technique has been termed 'maximum class discriminatory group (MCDG)'. The discrimination fraction works as the control parameter for the selection of bands. Subsequently, the unique bands are selected from the ranked DGs and used for HSI classification. The proposed MCDG technique has an articulate and quick implementation without requiring any initialization of bands and outperforms various state-of-the-art methods while experimenting using real HSI datasets. [ABSTRACT FROM AUTHOR]

As one of the most important techniques for hyperspectral image dimensionality reduction, band selection has received considerable attention, whereas self-representation subspace clustering-based band selection algorithms have received quite a lot of attention with good effect. However, many of them lack the self-supervision of representations and ignore the multi-level spectral–spatial information of HSI and the connectivity of subspaces. To this end, this paper proposes a novel self-supervised multi-level representation learning fusion-based maximum entropy subspace clustering (MLRLFMESC) method for hyperspectral band selection. Firstly, to learn multi-level spectral–spatial information, self-representation subspace clustering is embedded between the encoder layers of the deep-stacked convolutional autoencoder and its corresponding decoder layers, respectively, as multiple fully connected layers to achieve multi-level representation learning (MLRL). A new auxiliary task is constructed for multi-level representation learning and multi-level self-supervised training to improve its capability of representation. Then, a fusion model is designed to fuse the multi-level spectral–spatial information to obtain a more distinctive coefficient matrix for self-expression, where the maximum entropy regularization (MER) method is employed to promote connectivity and the uniform dense distribution of band elements in each subspace. Finally, subspace clustering is conducted to obtain the final band subset. Experiments have been conducted on three hyperspectral datasets, and the corresponding results show that the proposed MLRLFMESC algorithm significantly outperforms several other band selection methods in classification performance. [ABSTRACT FROM AUTHOR]

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. [ABSTRACT FROM AUTHOR]

This article provides insights into the optical signatures of plastic litter based on a published laboratory-scale reflectance data set (350–2500 nm) of dry and wet plastic debris under clear and turbid waters using different band selection techniques, including sparse variable selection, density peak clustering, and hierarchical clustering. The variable selection method identifies important wavelengths by minimizing a reconstruction error metric, while clustering approaches rely on the strengths of the correlation and local density of the spectra. Analyses of the data reveal three distinct absorption lines at 560, 740, and 980 nm that produce relatively broad reflectance peaks in the measured spectra of wet plastics around 475–490, 635–650, 810–815, and 1070 nm. The results of band selection consistently identify three important regions across 450–470, 650–690, and 1050–1100 nm that are close to the reflectance peaks of the mean of wet plastic spectra over clear and turbid waters. However, as the number of isolated important wavelengths increases, the results of the methodologies diverge. Density peak clustering identifies additional wavelengths in the short-wave infrared (SWIR) region of 1170–1180 nm) as a result of a high local density of the reflectance points. In contrast, hierarchical clustering isolates more wavelengths in the visible range of 365–400 nm due to weak correlations of nearby wavelengths. The results of the clustering methods are not consistent with the visual inspection of the signatures as peaks and valleys in the spectra, which are effectively captured by the variable selection method. It is also found that the presence of suspended sediments can (i) shift the important wavelength towards higher values in the visible part of the spectrum by less than 50 nm , (ii) attenuate the magnitude of wet plastic reflectance by up to 80% across the entire spectrum, and (iii) manifest a similar spectral signature with plastic litter from 1070 to 1100 nm. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Remote Sensing is the property of Editorial Office of Journal of Remote Sensing & Science Publishing Co. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Tactical aerial reconnaissance missions using small unmanned aerial systems (UASs) have become a common scenario in the military. In particular, the detection of visually obscured objects such as camouflage materials and unexploded ordnance (UXO) is of great interest. Hyperspectral sensors, which provide detailed spectral information beyond the visible spectrum, are highly suitable for this type of reconnaissance mission. However, the additional spectral information places higher demands on system architectures to achieve efficient and robust data processing and object detection. To overcome these challenges, the concept of data reduction by band selection is introduced. In this paper, a specialized and robust concept of context-based hyperspectral sensor management with an implemented methodology of band selection for small and challenging UXO and camouflaged material detection is presented and evaluated with two hyperspectral datasets. For this purpose, several anomaly detectors such as LRX, NCC, HDBSCAN, and bandpass filters are introduced as part of the detection workflows and tested together with the sensor management in different system architectures. The results demonstrate how sensor management can significantly improve the detection performance for UXO compared to using all sensor bands or statistically selected bands. Furthermore, the implemented detection workflows and architectures yield strong performance results and improve the anomaly detection accuracy significantly compared to common approaches of processing hyperspectral images with a single, specialized anomaly detector.

In environmental studies, rhodamine dyes are commonly used to trace water movements and pollutant dispersion. Remote sensing techniques offer a promising approach to detecting rhodamine and estimating its concentration, enhancing our understanding of water dynamics. However, research is needed to address more complex environments, particularly optically shallow waters, where bottom reflectance can significantly influence the spectral response of the rhodamine. Therefore, this study proposes a novel approach: transferring pre-trained classifiers to develop a generalizable method across different environmental conditions without the need for in situ calibration. Various samples incorporating distilled and seawater on light and dark backgrounds were analyzed. Spectral analysis identified critical detection regions (400–500 nm and 550–650 nm) for estimating rhodamine concentration. Significant spectral variations were observed between light and dark backgrounds, highlighting the necessity for precise background characterization in shallow waters. Enhanced by the Sequential Feature Selector, classification models achieved robust accuracy (>90%) in distinguishing rhodamine concentrations, particularly effective under controlled laboratory conditions. While band transfer was successful (>80%), the transfer of pre-trained models posed a challenge. Strategies such as combining diverse sample sets and applying the first derivative prevent overfitting and improved model generalizability, surpassing 85% accuracy across three of the four scenarios. Therefore, the methodology provides us with a generalizable classifier that can be used across various scenarios without requiring recalibration. Future research aims to expand dataset variability and enhance model applicability across diverse environmental conditions, thereby advancing remote sensing capabilities in water dynamics, environmental monitoring and pollution control.

Band selection (BS) aims to reduce redundancy in hyperspectral imagery (HSI). Existing BS approaches typically model HSI only in a single dimension, either spectral or spatial, without exploring the interactions between different dimensions. To this end, we propose an unsupervised BS method based on a spectral–spatial cross-dimensional attention network, named SSANet-BS. This network is comprised of three stages: a band attention module (BAM) that employs an attention mechanism to adaptively identify and select highly significant bands; two parallel spectral–spatial attention modules (SSAMs), which fuse complex spectral–spatial structural information across dimensions in HSI; a multi-scale reconstruction network that learns spectral–spatial nonlinear dependencies in the SSAM-fusion image at various scales and guides the BAM weights to automatically converge to the target bands via backpropagation. The three-stage structure of SSANet-BS enables the BAM weights to fully represent the saliency of the bands, thereby valuable bands are obtained automatically. Experimental results on four real hyperspectral datasets demonstrate the effectiveness of SSANet-BS.

Band selection (BS) work for hyperspectral remote sensing images (HRSIs) has attracted increasing attention from researchers. In the HRSI, the salient features of each band always include the more important discriminative features. However, most existing BS methods always have difficulty in simultaneously performing local-global consistency analysis on salient features in bands of HRSIs. In addition, these methods also ignore the consideration of the diversity of band information to some extent when selecting a band subset. To tackle these problems, a novel unsupervised hyperspectral BS method combining the variation degree of salient features (UBSF) is proposed in this article. UBSF first promotes the singular value decomposition method to analyse the salient features in bands and develops a new low-dimensional local-global salient feature representation (NLR) method for discriminative feature representations of the bands. The NLR method combines the variation degree of the features of the significant representative patches of each band in their corresponding significant feature subspaces to achieve a localglobal consistency analysis of salient features of each band of HRSIs. Next, UBSF proposes a new two-stage strategy (TSS) to preserve the diversity of BS results as much as possible. In TSS, we propose a novel similarity criterion, which can make the highly similar band subsets with near information amount in the overall bands adaptively compressed according to the obtained similarity results of bands further. Then, the band subset with the most information is selected as the final BS results in the obtained reduceddimensional HRSI. Finally, the validity of the UBSF method is verified based on three measured HRSI datasets. The experimental results show that the proposed UBSF is superior to other state-of-the-art BS methods. [ABSTRACT FROM AUTHOR]

In the domain of real-world concept learning, feature selection plays a crucial role in accelerating learning processes and enhancing the quality of classification concepts, particularly in remote sensing image classification. Traditional methods often exhibit weaknesses such as instability, sensitivity to noise, and a lack of consideration for feature dependencies when selecting informative spectral features from hyper-spectral (HS) images. Recent studies have recommended the utilization of meta-heuristic algorithms to address these limitations. However, most of these approaches overlook the correlation between spectral bands. To tackle these challenges, we propose a novel migration-based particle swarm optimization (MBPSO) algorithm that surpasses previous methods in terms of exploitation and exploration and considers spectral band correlation to prevent the inclusion of unnecessary bands. The variance-based J1 criteria are employed as the fitness function for MBPSO, while a support vector machine (SVM) is utilized as the classifier. To evaluate the performance of our proposed method against other optimization algorithms, we conducted experiments using four HS datasets, namely Indian Pine, Pavia University, Salinas, and Washington Mall. Numerical evaluations encompassing class accuracy, overall accuracy, kappa coefficient, execution time, stability, Wilcoxon test, and convergence rate were performed. The results indicate that our proposed algorithm achieves superior numerical outcomes compared to other algorithms, while requiring fewer features to attain higher accuracy. The algorithm's higher speed, reduced computational time, and avoidance of additional features in the solution can be attributed to the presence of the conditional mutation operator within the MBPSO algorithm. [ABSTRACT FROM AUTHOR]

The increasing number of space missions have resulted in an augmented density of artificial objects positioned in orbital domains near Earth. Knowing the material composition of the resident space object can be of value in object identification and risk assessment using remote sensing techniques. To perform this task, it is increasingly imperative to optically characterize spacecraft materials to identify unique material-specific spectroscopic markers. In this work, a variety of materials frequently utilized by the aerospace industry in spacecraft design and construction were analyzed using reflectance spectroscopy. The collected data provide a spectral characterization baseline for modern-day and historical spacecraft materials. The color index was computed for standard astronomical and suggested theoretical filter passbands and compared to each other in their capability to discriminate materials belonging to different classes. The color index was calculated from reflectance spectra of common spacecraft materials in their pristine, as-received conditions that fell under different family groupings. Different color-index combinations were studied using color-color diagrams. Visual and quantitative analysis of the color-color diagrams were used to evaluate the possibility of discriminating materials from one another by means of optical measurements. Results of the analysis show that polyimide and photovoltaic materials are most easy to discriminate from all other materials via color indices using different filter passbands evaluated in this study. [ABSTRACT FROM AUTHOR]