Marine mucilage outbreaks not only pose a threat to the marine ecosystems, but also are a detriment to economy and public-health. The recent marine mucilage outbreak of Spring 2021 in the Sea of Marmara, Türkiye, was one of most serious recorded mucilage outbreaks, lasting over three months and covering more than 1000 square kilometers. Recently, linear spectral unmixing based environmental monitoring of marine mucilage from hyperspectral data have been shown to provide easy to interpret analysis of this complex phenomenon, in terms of endmember signatures and fractional abundances. This paper carries the work forward and proposes nonlinear unmixing for the environmental monitoring of marine mucilage. Hyperspectral data acquired by the PRISMA mission are used in this investigation.

Disasters in marine ecosystems, such as the outbreak of mucilage in the inland Sea of Marmara in the spring of 2021, raise serious environmental, economic and public health concerns. Recently, methods based on fully unsupervised unmixing and dealing with spectral variability have made it possible to analyze marine mucilage, quantify its abundance and thus to map this harmful phenomenon. This work proposes to use an informed (or semi-supervised) unmixing technique, which is based on nonnegative matrix factorization and dealing with spectral variability, using a field-measured spectrum of high-density (or accumulated) mucilage, in order to detect and map, in particular, this marine material. The used technique is evaluated on real hyperspectral PRISMA data, and compared with other unmixing-based techniques.

The marine mucilage outbreak in the Sea of Marmara in Spring 2021 sparked renewed interest in this complex environmental phenomenon and in monitoring the delicate marine ecosystems of our planet. A number of works in the literature have utilized remote sensing spectral indices with and without classification-based approaches in order to detect and map marine mucilage for this outbreak. Recently, SHAP method, often utilized for the purpose of Explainable Artificial intelligence (XAI) has been utilized to highlight that water-related spectral indices have varying contributions in the detection of marine mucilage. This work continues that line of research in incorporating a large number of water, soil, vegetation, and mucilage indices. Preliminary results, presented in this paper, partially support the findings in the literature while still presenting some interesting outcomes and insights.

Earth observation (EO) sensors and remote sensing play a crucial role in the detection and analysis of environmental hazards. Among these hazards, monitoring and understanding marine pollution remain challenging problems due to the complex and dynamic spatio-temporal characteristics of pollution and water. The recent outbreak of marine mucilage (or sea-snot) in the inland Sea of Marmara in the Spring of 2021 is a striking example of a dynamic environmental pollution, This work investigates temporal analysis of mucilage by unmixing based change detection on temporal hyperspectral satellite data acquired with the recently launched PRISMA sensor. The proposed unmixing-based approach enables detecting temporal changes in terms of endmembers and abundances, and hence provides information on the nature of the change, in an unsupervised manner without any training step.

This study addresses the explainability challenges of deep-learning models in the context of hyperspectral remote sensing image classification. Three prominent explainable artificial intelligence methods, namely GradCAM, GradCAM++, and Guided Backpropagation, have been employed in order to comprehend the decision-making process of a typical convolutional neural network model during spatial-spectral hyperspectral image classification. The experiments that have been conducted investigate the impact of pixel patch sizes on spatial attention, as well as spectral band importance. The findings provide insights into the behavior of both convolutional neural networks, as well as the comparative performance of explainability techniques.

The mucilage outbreak in the Sea of Marmara in the spring of 2021 has once again emphasized the importance of addressing climate and pollution associated hazards. Although multispectral images have traditionally been used for such purposes, an analysis of marine mucilage, with its spectral similarity to marine debris, and its spectral variations due to composition and/or sediment or bacterial aggregation, is a prime candidate to benefit from the advantages of hyperspectral data. The recently launched PRISMA mission provides an important opportunity to this end. This work proposes the use of unmixing on PRISMA datasets in order to analyze the spectral characteristics, the variation due to aggregation, and the spatial distribution, of marine mucilage. The proposed approach provides consistent and relevant information on two different datasets, with the potential to benefit cleaning and understanding efforts for marine mucilage. In addition, unlike the previous studies with supervised classification, the proposed approach does not require a training step, and the abundance fraction maps obtained using unmixing are easy to interpret and analyze for mucilage aggregation.

The latest sea-snot, i.e. mucilage, outbreak in the Sea of Marmara hit all the headlines in Turkey in the Spring-Summer of 2021. Its slimy mucus characteristic was seen on the sea surface, but marine researchers warned that it spread down to 30 metres below the surface and could cause serious water-borne diseases, in addition to its detriment to the economy. Prevention and clean-up measures have been started and are ongoing. In this context, using remote sensing approaches can provide a significant advantage for understanding not only its spatial distribution throughout the Sea of Marmara but also its spectral and biochemical properties. In this work, the ag-gregation and spatial distribution of mucilage are investigated in the Sea of Marmara, Turkey, using hyperspectral data acquired by the PRISMA sensor.

This paper presents an unmixing based change detection (UBCD) approach based on constrained nonnegative matrix factorization (NMF) for hyperspectral images. UBCD provides not only multi-output change detection, but also subpixel level information about the nature of the changes that occur in the scene. The proposed method utilizes constrained NMF with the sparsity constraint for the abundances and the minimum volume constraint for the endmembers, reducing the solution space for the matrix factorization and resulting in enhanced unmixing and change detection performance. The change detection output is obtained in terms of the temporal abundance matrix differences for each endmember. The proposed method is evaluated on synthetic and real multitemporal datasets.

With the increase in the number and availability of imaging sensors, data size or dimensionality reduction, compression, archiving and retrieval algorithms are gaining importance. Whereas in the past, for hyperspectral images, the size reduction has been mostly concerned with the spectral domain, the increase in the number and spatial sizes of the hyperspectral images has raised the question whether a spatial reduction is also feasible. However, this reduction should be conducted in a content-aware fashion and should preserve the important and relevant information in the scene whether for archiving and retrieval purposes, or for following image processing tasks. In this work, an anomaly-preserving content-aware size reduction approach is proposed for hyperspectral images. The approaches utilizes seam carving with a spatial homogeneity energy function to preserve anomalies while performing reduction. Synthetic and real datasets are used to validate the proposed methodology.

Pansharpening is the fusion of panchromatic (PAN) image and multispectral (MS) or hyperspectral (HS) images and provides high spatial and high spectral resolution MS or HS images. Pansharpening mainly extracs the high frequency details from the PAN image, and then injects these details to the MS or HS image. This detail injection procedure can be performed in a variety of ways: using global, block-based or clustering-based techniques. In this paper, block-based and clustering-based approaches are utilized for standart component substitution pansharpening approaches, namely Intensity Hue Saturation (IHS), Brovey Transform (BT), Gram Schmidt (GS) orthagonalization procedure and Principal Component Analysis (PCA) techniques. Both nonoverlapping and overlapping blocks are considered, along with various segmentation approaches such as k-means, Iterative Self Organizing Data Analysis Techniques Algorithm (ISODATA) and Simple Linear Iterative Clustering (SLIC). Two datasets with different characteristics are used in order to evaluate the approaches, and the block-based and segmentation-based approaches are shown to provide enhanced performance.

Data dimensionality reduction is crucial for hyperspectral imaging, whether to address the Hughes phenomenon or decrease the computational cost and memory requirements. Whereas up until recently, the majority of dimensionality reduction approaches for hyperspectral images have operated in the spectral domain, the spatial sizes of the hyperspectral images have also been increasing, and as a result, reduction or compression of hyperspectral images in the spatial domain is also gaining importance. However, spatial size reduction is often problematic as it heavily depends on the contents of the image in question. In this paper, seam carving, a content-aware image size reduction approach, is adapted to hyperspectral images for spatial size reduction. The approach is evaluated using multiple energy functions, and the potential and advantages of content-aware size reduction in the spatial domain for hyperspectral images is presented and validated through the example application of unmixing.

Until recently, hyperspectral image denoising was considered as a prior step to applications such as classification, detection, or unmixing. However, unmixing has been recently shown to also provide denoising due to its inherent property of representing pixels in terms of pure material signatures and their abundances. It is possible to eliminate sensor-induced or atmospheric noise by unmixing based denoising, by not including these noise effects in the endmember signatures. Up until now, only spectral unmixing, and in a more recent paper spectral unmixing after spatial preprocessing, have been utilized for hyperspectral denoising. This letter proposes the use of spatial - spectral sparse unmixing for hyperspectral denoising. Sparse unmixing has the advantage of circumventing dimensionality detection, while the use of spatial processing in the sparse regression further enhances the unmixing and denoising performance. The proposed approach provides enhanced denoising and inpainting performance with respect to previously proposed unmixing based change detection approaches.

Change detection by unmixing has been shown to provide enhanced change detection performance for hyperspectral images with respect to more traditional approaches, especially when the temporal images contain sub-pixel level changes. In a recent paper, change detection by spectral unmixing was investigated in detail and the advantages that can be gained by using such an approach were systematically presented through various experimental studies. However, the utilized unmixing-based change detection approach relied solely on spectral information and disregarded the spatial distribution in the scene, which inevitably limits the performance that can be achieved. In this paper, superpixels are used to integrate the spatial information in the image into the unmixing process, which in turn enhances the change detection performance with respect to spectral unmixing based change detection.

Applying spectral unmixing on a series of multitemporal hyperspectral images for change detection has the potential to reveal important subpixel-level information, such as the abundance variation of each underlying material in a given location or the change in the distribution of materials throughout the scene, with time or resulting from significant events such as a natural disaster. However, change detection by spectral unmixing for hyperspectral images has not been extensively studied up to now, and most studies have been limited to specific cases and data sets. This is caused by the scarcity of real multitemporal hyperspectral data and the inherent difficulties in applying unmixing to multitemporal hyperspectral data in a coherent way. In this letter, we investigate change detection for hyperspectral images by spectral unmixing and systematically present the advantages that can be gained by using such an approach, supported by experimental studies conducted on carefully prepared synthetic data sets and also with real data sets.

Hyperspectral imaging provides increased capability for many image processing tasks with respect to standard imaging systems. One of such tasks in hyperspectral image processing is change detection, which aims to detect the differences occurring between images acquired from the same scene at different times. In this paper, a panoramic hyperspectral imaging system is used to capture multitemporal hyperspectral data, and novel multi-band Census Transform (MCT) is proposed for change detection on these data. Experimental results validate the performance of the proposed method for the utilized acquisition system.

Hyperspectral imaging provides high spectral resolution and thereby improved classification, detection, and recognition capabilities with respect to standard imaging systems. However, hyperspectral images generally have low spatial resolution, varying from a few to tens of meters, resulting from technical limitations such as platform data storing capacity and satellite-to-ground transmission bandwidth. Spectral unmixing provides information on pixels in terms of abundances of pure spectral signatures, without providing spatial distribution at subpixel level. Multisensor image fusion approaches can provide such information but require an additional image with higher spatial resolution that is acquired in similar conditions with the hyperspectral image. In this letter, a novel spatial resolution enhancement method using fully constrained least squares (FCLS) spectral unmixing and spatial regularization based on modified binary particle swarm optimization is proposed to achieve spatial resolution enhancement in hyperspectral images, without using an additional image with higher spatial resolution. The proposed method has a highly parallel nature with respect to its counterparts in the literature and is fit to be adapted to field-programmable gate array architecture.

Spectral unmixing is the process of identifying pure spectral signatures, called endmembers, from a hyperspectral data, and then expressing each pixel vector in terms of the fractional abundances of these endmembers. Most of the endmember extraction methods in the literature use only the spectral information, whereas the spatial composition of the data is disregarded. Spatial preprocessing methods, that are motivated by the assumption that endmembers are more likely to be located in homogeneous regions instead of transition areas, can alleviate this drawback and hence increase the performance. However, such a preprocessing approach generally results in a failure of extracting anomalous endmembers which can be of importance for many applications. In this paper, a preprocessing approach that guides the endmember extraction process to homogenous regions while retaining the anomaly points, by combining spatial preprocessing with anomaly detection, is proposed.

This paper proposes to use empirical mode decomposition (EMD) with spectral gradient enhancement to increase the classification accuracy of hyperspectral images with support vector machine (SVM) classification. Recently, it has been shown that higher hyperspectral image classification accuracy can be achieved by using 2-D EMD that is applied to each hyperspectral band separately to obtain the intrinsic mode functions (IMFs) of each band, while the sum of the IMFs are used as feature data in the SVM classification process. In the previous approach, IMFs have been summed directly, i.e., with equal weights. It is shown in this paper, that it is possible to significantly increase the classification accuracy by using appropriate weights for the IMFs in the summation process. In the proposed approach, the weights of the IMFs are obtained so as to optimize the total absolute spectral gradient, and a genetic algorithm-based optimization strategy has been adopted to obtain the weights automatically in this way. While the 2-D EMD basically provides spatial processing, the proposed method further incorporates spectral enhancement into the process. It is shown that a significant increase in hyperspectral image classification accuracy can be achieved using the proposed approach.

This paper proposes an empirical mode decomposition (EMD) based approach with spectral gradient enhancement for hyperspectral image classification using support vector machines (SVM). In a previous study, it has been shown that using the sum of intrinsic mode functions (IMFs), obtained by applying two-dimensional (2D) EMD to each hyperspectral band, increases the classification accuracies significantly. In this paper, it is shown that using optimum weights for the IMFs, instead of the equal weight approach of the previous study, results in increased classification accuracies. The weights for the IMFs are obtained by a genetic algorithm (GA) based optimization strategy which aims to maximize spectral gradient and hence incorporate spectral processing with the spatial processing of 2D EMD.

In this study we propose an automated fine registration of EO-1 Hyperion and IKONOS imagery. An intensity based registration that is area-based and pixelwise is performed to register given images of divergent spatial and spectral resolution. Two similarity measures that are commonplace in image registration; NCC and NMI, and an operation that is particular to image restoration; CTO is adopted as an error measure as a novelty in image registration. We are convinced with the performance and efficiency of CTO compared to other two common methods of intensity-based registration.

This paper presents a comparison of the classification performance of some vector machine based classification methods, namely, Import Vector Machines (IVM), Support Vector Machines (SVM) and Relevance Vector Machines (RVM), for hyperspectral images. Evaluation is carried out in terms of the number of vectors and classification accuracies. Furthermore, novel to this paper, Discriminative Random Field method with Graph Cut algorithm is applied to the probabilistic classification output of IVM based hyperspectral classification results, and it is shown that this approach significantly increases classification accuracies.

This letter presents hyperspectral image segmentation based on the phase-correlation measure of subsampled hyperspectral data, which is referred to as modified phase correlation. The hyperspectral spectrum of each pixel is initially subsampled to gain robustness against noise and spatial variability, and phase correlation is applied to determine spectral similarity. Similar and dissimilar pixels are decided according to the peak value of the phase correlation result to determine pixels that fall into the same segments. The approach can be regarded as a region-growing technique. The total number of segments is determined automatically according to the similarity threshold

One-bit transforms (1BTs) have been proposed for low-complexity block-based motion estimation by reducing the representation order to a single bit, and employing binary matching criteria. However, as a single bit is used in the representation of image frames, bad motion vectors are likely to be resolved in 1BT-based motion estimation algorithms particularly for small block sizes. It is proposed in this paper to utilize a two-bit transform (2BT) for block-based motion estimation. Image frames are converted into two-bit representations by a simple block-by-block two bit transform based on multithresholding with mean and linearly approximated standard deviation values. In order to avoid blocking effects at block boundaries during the block-by-block transformation while enabling the two-bit representation to be constructed according to local detail, threshold values are computed within a larger window surrounding the transforming block. The 2BT makes use of lower bit-depth and binary matching criteria properties of 1BTs to achieve low-complexity block motion estimation. The 2BT improves motion estimation accuracy and seriously reduces the amount of bad motion vectors compared to 1BTs, particularly for small block sizes. It is shown that the proposed 2BT-based motion estimation technique improves motion estimation accuracy in terms of peak signal-to-noise ratio of reconstructed frames and also results in visually more accurate frames subsequent to motion compensation compared to the 1BT-based motion estimation approach.