Search

Cosmogenic nuclide exposure dating is an important technique for reconstructing glacial histories. Many of the most commonly applied cosmogenic nuclides are extracted from the mineral quartz, meaning sampling of felsic (silica-rich) rock is often preferred to sampling of mafic (silica-poor) rock for exposure dating studies. Fieldwork in remote regions such as Antarctica is subject to time constraints and considerable logistical challenges, making efficient sample recovery critical to successful research efforts. Remote sensing offers an effective way to map the geology of large areas prior to fieldwork and expedite the sampling process. In this study, we assess the viability of multispectral remote sensing to distinguish felsic from mafic rock outcrops at visible-near infrared (VNIR) and shortwave infrared (SWIR) wavelengths using both the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) and very high-resolution Worldview-3 (WV-3) imagery. We applied a combination of spectral mapping and ground truth from spectral measurements of 17 rock samples from Mount Murphy in the Amundsen Sea sector of West Antarctica. Using this approach, we identified four dominant rock types which we used as a basis for felsic–mafic differentiation: felsic granites and gneisses, and mafic basalts and fragmental hydrovolcanic rocks. Supervised classification results indicate WV-3 performs well at differentiating felsic and mafic rock types and that ASTER, while coarser, could also achieve satisfactory results and be used in concert with more targeted WV-3 image acquisitions. Finally, we present a revised felsic–mafic geological map for Mt Murphy. Overall, our results highlight the potential of spectral mapping for preliminary reconnaissance when planning future cosmogenic nuclide sampling campaigns in remote, unvisited areas of the polar regions. [ABSTRACT FROM AUTHOR]

Space-borne hyperspectral imagery data are known for their high spectral resolution in a number of narrow wavelength intervals, which makes these data useful for mineral mapping. However, the available free-of-charge hyperspectral scenes cover only narrow and scattered geographic areas. In contrast, multispectral imagery scenes have a nearly complete spatial coverage and wider wavelength intervals. The low spectral resolution of the multispectral data, however, limits their efficiency in the mineral mapping of small geological massifs or hydrothermal alteration halos. The present contribution presents a new transformation tool (SAM-HIT) to simulate the hyperspectral sensor responses in unscanned areas based on partially overlapping hyperspectral and multispectral scenes. Simulation or prediction of the pseudohyperspectral data is here accomplished by using the simulated annealing linear optimization algorithm, which allows the lowest possible mismatch between the original and predicted data. The high visual and numerical correlation of the resultant data confirms the reliability of the newly adopted transformation. Further, the application of the SAM-HIT to a well-exposed part of the Egyptian basement complex with available hyperspectral data showed high concordance and nearly identical band signatures, opening a new outlook for mineral exploration in vast areas by a nearly automated cost-free means. [ABSTRACT FROM AUTHOR]

In this article, we have processed Landsat-8 spectral bands, ground gravity, and magnetic data for delineating the mafic cumulates within the anorthositic-layered complex in Sittampundi, Tamilnadu, India. In this regard, surface anomaly map of mafic cumulates was derived by processing spectral bands of Operational Land Imager (OLI) sensor using matched filtering method. Residual gravity, magnetic anomaly maps are also derived. Spatial position of geophysical anomalies corroborate well with the spectral anomalies. A promising zone of chromitite occurrences is identified at the eastern fringe of the study area based on the concurrence of spectral and geophysical anomalies. Spatial profiles of geophysical and spectral anomalies around this promising zone are similar to the respective spatial profiles observed around the known chromite bearing mafic cumulates. Depth continuity of the mafic cumulates is confirmed by deriving of gravity and magnetic anomaly map at different depth using upward continuation filters to confirm concealed deposit. [ABSTRACT FROM AUTHOR]

Abstract We searched for an optimized protocol for mapping observations from a point spectrometer onto a shape model composed of triangular facets, in the context of NASA's asteroid sample return mission, OSIRIS‐REx (Origins, Spectral Interpretation, Resource Identification, and Security‐Regolith Explorer). Our study was conducted before the spacecraft arrived at the mission target asteroid (101955) Bennu, and we used observational sequence plans of the OSIRIS‐REx Visible and InfraRed Spectrometer (OVIRS). We explored six methods of mapping data to shape model facets, using three spatial resolutions. We attempted to boost map fidelity by increasing the observational coverage of the surface. We find that increasing shape model resolution improves mapping quality. However, once the shape model's mean facet edge length is smaller than two‐fifths of the diameter of the instrument's field of view (FOV), the increase in quality tapers off. The six mapping methods can be broken into two categories: facets that (1) select or (2) combine (average) data from observations. The quality differences between similar averaging methods (clipped average, weighted average, etc.) are insignificant. Selecting the nearest observation to a facet best preserves an enclosed outcrop's shape and signal, but averaging spots are more conservative against errors in photometric modeling. A completely enclosed outcrop border expands into the surrounding region by 0.8–1.5 radii of the instrument's FOV. Regions smaller than the instrument's FOV are present in resulting maps; however, their signal strength is reduced as a function of their size relative to the instrument FOV.

We searched for an optimized protocol for mapping observations from a point spectrometer onto a shape model composed of triangular facets, in the context of NASA's asteroid sample return mission, OSIRIS‐REx (Origins, Spectral Interpretation, Resource Identification, and Security‐Regolith Explorer). Our study was conducted before the spacecraft arrived at the mission target asteroid (101955) Bennu, and we used observational sequence plans of the OSIRIS‐REx Visible and InfraRed Spectrometer (OVIRS). We explored six methods of mapping data to shape model facets, using three spatial resolutions. We attempted to boost map fidelity by increasing the observational coverage of the surface. We find that increasing shape model resolution improves mapping quality. However, once the shape model's mean facet edge length is smaller than two‐fifths of the diameter of the instrument's field of view (FOV), the increase in quality tapers off. The six mapping methods can be broken into two categories: facets that (1) select or (2) combine (average) data from observations. The quality differences between similar averaging methods (clipped average, weighted average, etc.) are insignificant. Selecting the nearest observation to a facet best preserves an enclosed outcrop's shape and signal, but averaging spots are more conservative against errors in photometric modeling. A completely enclosed outcrop border expands into the surrounding region by 0.8–1.5 radii of the instrument's FOV. Regions smaller than the instrument's FOV are present in resulting maps; however, their signal strength is reduced as a function of their size relative to the instrument FOV. Key Points: Spectral map quality is controlled by shape model facet size, boresight spot size, and method of assigning spot data to facetsShape models with facet edges smaller than 2/5 the boresight spot diameter cease to improve map fidelityData selection algorithms perform better than data combination methods for mapping distinct outcrops [ABSTRACT FROM AUTHOR]

In this paper, we present a novel framework for a voice conversion (VC) system based on a cyclic recurrent neural network (CycleRNN) and a finely tuned WaveNet vocoder. Even though WaveNet is capable of producing natural speech waveforms when fed with natural speech features, it still suffers from speech quality degradation when fed with oversmoothed features, such as spectral parameters estimated from a statistical model. One way to address this problem is to introduce oversmoothed features while developing a WaveNet model. However, in a VC framework, providing oversmoothed spectral features of a target speaker for WaveNet modeling is not straightforward owing to the difference in the time-sequence alignment from that of a source speaker. To overcome this problem, we propose the use of a cyclic spectral conversion network, i.e., CycleRNN, capable of performing a conversion flow, i.e., source-to-target, and a cyclic flow, i.e., to generate self-predicted target spectra. The CycleRNN spectral model is trained using both conversion and weighted cyclic losses. To finely tune WaveNet, a pretrained multispeaker WaveNet model is optimized using the self-predicted features of the corresponding target speaker of a speaker conversion pair. The experimental results demonstrate that 1) the proposed CycleRNN-based spectral model for WaveNet fine-tuning significantly improves the naturalness of the converted speech waveforms, giving an overall mean opinion score of 3.50; and 2) the proposed model yields the highest speaker conversion accuracy, giving an overall speaker similarity score of 78.33%, which is a significant improvement compared with conventional WaveNet fine-tuning using natural target features.

Exploring techniques that breakthrough the unknown space or material species is of considerable significance to military and civilian fields, and it is a challenging task without any prior information. Nowadays, the use of material-specific spectral information to detect unknowns has received increasing interest. However, affected by noise and interference, high-dimensional hyperspectral anomaly detection is difficult to meet the requirements of high detection accuracy and low false alarm rate. Besides, there is a problem of insufficient and unbalanced samples. To address these problems, we propose a novel hyperspectral anomaly detection framework based on spectral mapping and feature selection (SMFS) in an unsupervised manner. The SMFS introduces the essential properties of hyperspectral data into an unsupervised neural network to construct the nonlinear mapping relationship from high-dimensional spectral space to low-dimensional deep feature space. And it searches the optimal feature subset from the candidate feature space for standing out anomalies. Because of the compelling characterization of the encoder, we develop it specifically for spectral signatures to reveal the hidden data. Quantitative and qualitative experiments on real hyperspectral datasets indicate that the proposed method can provide the compact features overcoming the problems of noise, interference, redundancy and time-consuming caused by high-dimensionality and limited samples. And it has advantages over some state-of-the-art competitors concerning detecting anomalies of different scales. [ABSTRACT FROM AUTHOR]

This paper presents an evaluation of parallel voice conversion (VC) with neural network (NN)-based statistical models for spectral mapping and waveform generation. The NN-based architectures for spectral mapping include deep NN (DNN), deep mixture density network (DMDN), and recurrent NN (RNN) models. WaveNet (WN) vocoder is employed as a high-quality NN-based waveform generation. In VC, though, owing to the oversmoothed characteristics of estimated speech parameters, quality degradation still occurs. To address this problem, we utilize post-conversion for the converted features based on direct waveform modifferential and global variance postfilter. To preserve the consistency with the post-conversion, we further propose a spectrum differential loss for the spectral modeling. The experimental results demonstrate that: (1) the RNN-based spectral modeling achieves higher accuracy with a faster convergence rate and better generalization compared to the DNN-/DMDN-based models; (2) the RNN-based spectral modeling is also capable of producing less oversmoothed spectral trajectory; (3) the use of proposed spectrum differential loss improves the performance in the same-gender conversions; and (4) the proposed post-conversion on converted features for the WN vocoder in VC yields the best performance in both naturalness and speaker similarity compared to the conventional use of WN vocoder. [ABSTRACT FROM AUTHOR]

The emergence of functional magnetic resonance imaging (fMRI) provides a good opportunity for brain functional parcellation. However, the high dimension and low signal-to-noise ratio of fMRI data brings difficulties to the existing parcellation methods. To address the issue, this paper presents a novel brain functional parcellation method based on artificial bee colony clustering (ABCC) algorithm with self-adaptive crossover and stepwise search (called CSABCC). In CSABCC, the preprocessed fMRI data is first mapped into a low-dimensional space by spectral mapping to reduce its dimension and each food source position is encoded as a clustering solution composed of cluster centers. Then, CSABCC utilizes an improved artificial bee colony search procedure with some robustness advantage to seek better food sources, where a self-adaptive crossover is employed to enhance information exchange between individuals and onlooker bees adopt a stepwise search to improve its search capability. Finally, a functional parcellation result is obtained by mapping cluster labels onto the corresponding voxels. The experiments on simulated fMRI data show that CSABCC can generate the parcellation closest to the real result, and these results on real insula fMRI data also demonstrate that CSABCC has better search capability and can produce parcellation structures with stronger functional consistency and regional continuity compared to some other typical algorithms. Moreover, the correctness of the parcellation results is also validated by functional connectivity fingerprints of the corresponding subregions. [ABSTRACT FROM AUTHOR]

Abstract: A concise insight into the outputs provided by the latest prototype of visible‐near infrared (Vis‐NIR) multispectral scanner (National Research Council‐National Institute of Optics, CNR‐INO, Italy) is presented. The analytical data acquired on an oil painting Madonna of the Rabbit by É. Manet are described. In this work, the Vis‐NIR was complemented with X‐ray fluorescence (XRF) mapping for the chemical and spatial characterization of several pigments. The spatially registered Vis‐NIR data facilitated their processing by spectral correlation mapping (SCM) and artificial neural network (ANN) algorithm, respectively, for pigment mapping and improved visibility of pentimenti and of underdrawing style. The data provided several key elements for the comparison with a homonymous original work by Titian studied within the ARCHive LABoratory (ARCHLAB) transnational access project. [ABSTRACT FROM AUTHOR]

Natural product screening programs have uncovered molecules from diverse natural sources with various biological activities and unique structures. However, much is yet underexplored and additional information is hidden in these exceptional collections. We applied untargeted mass spectrometry approaches to capture the chemical space and dispersal patterns of metabolites from an in-house library of marine cyanobacterial and algal collections. Remarkably, 86% of the metabolomics signals detected were not found in other available datasets of similar nature, supporting the hypothesis that marine cyanobacteria and algae possess distinctive metabolomes. The data were plotted onto a world map representing eight major sampling sites, and revealed potential geographic locations with high chemical diversity. We demonstrate the use of these inventories as a tool to explore the diversity and distribution of natural products. Finally, we utilized this tool to guide the isolation of a new cyclic lipopeptide, yuvalamide A, from a marine cyanobacterium.

Fusion of the high-spatial-resolution hyperspectral (HHS) image using low-spatial- resolution hyperspectral (LHS) and high-spatial-resolution multispectral (HMS) image is usually formulated as a spatial super-resolution problem of LHS image with the help of an HMS image, and that may result in the loss of detailed structural information. Facing the above problem, the fusion of HMS with LHS image is formulated as a nonlinear spectral mapping from an HMS to HHS image with the help of an LHS image, and a novel cluster-based fusion method using multi-branch BP neural networks (named CF-BPNNs) is proposed, to ensure a more reasonable spectral mapping for each cluster. In the training stage, considering the intrinsic characteristics that the spectra are more similar within each cluster than that between clusters and so do the corresponding spectral mapping, an unsupervised clustering is used to divide the spectra of the down-sampled HMS image (marked as LMS) into several clusters according to spectral correlation. Then, the spectrum-pairs from the clustered LMS image and the corresponding LHS image are used to train multi-branch BP neural networks (BPNNs), to establish the nonlinear spectral mapping for each cluster. In the fusion stage, a supervised clustering is used to group the spectra of HMS image into the clusters determined during the training stage, and the final HHS image is reconstructed from the clustered HMS image using the trained multi-branch BPNNs accordingly. Comparison results with the related state-of-the-art methods demonstrate that our proposed method achieves a better fusion quality both in spatial and spectral domains.

Arctic environments provide a challenging ground for geological mapping and mineral exploration. Inaccessibility complicates ground surveys and the presence of ice, vegetation, and lichens hinders supportive remote sensing surveys. Spectral mixing of lichens and bare rock can shift the wavelength position of characteristic absorption features, thereby complicating the spectral mapping of minerals and lithologies. The extent to which diagnostic rock features are preserved despite the presence of lichens is of major concern in remotely sensed geological studies and estimates of the critical level of lichen coverage, below which spectral features of the mineral substrate can still be identified, are needed. We investigated how surficial lichen cover affects the characteristics of shortwave infrared (SWIR) mineral absorption features and the efficacy of automated absorption feature extraction. For this purpose, mixed spectra were synthetically generated from laboratory spectra of common rock forming SWIR absorbing minerals and lichens. Wavelength displacements of characteristic absorption features for each mixed spectrum were then analyzed as a function of percentage lichen cover. Distinctive trends were identified that can be used in future analysis: The strong spectral features of mica group minerals around 2200 and 2340–2350 nm maintain their integrity for up to 30% lichen cover, despite the related shift toward shorter wavelengths for higher percentage lichen cover. In contrast, very weak absorption bands around 2440 nm in (white) micas spectra are completely obscured for a lichen cover of ≥ 50%. Our observations suggest that the chlorite feature around 2250 nm is shifted toward longer wavelengths and the depth of this feature as well as the contrast between lichen and substrate spectra define the amount of lichen needed to mask it. Furthermore, lichens induce a spectral shift towards shorter wavelengths for the features around 2320 nm for the rocks containing amphibole, chlorite, carbonate and serpentine group minerals. In addition, no wavelength displacement is observed for chlorite, biotite and phlogopite features around 2380 nm in mixtures with lichens. By quantifying lichen cover effects on mineral absorption features, our study highlights the importance of being precautious in any interpretation in areas characterized by abundant lichen-covered outcrops. This can be of significant importance for mineral and deposit vectoring as the presence of abundant lichen coverage causing slightly shifted features for a given spectra can be erroneously identified as a path to a deposit. [ABSTRACT FROM AUTHOR]

Hyperspectral imaging (HSI) techniques using a newly-developed frame camera were applied to determine internal properties of mango fruits including firmness, total soluble solids (TSS) and titratable acidity (TA). Prediction models were developed using spectral data from relative surface reflectance of 160 fruits in the visible and near infrared (vis/NIR) region of 450–998 nm analysed by PLS regression. For data reduction, MLR analysis showed 16 significant factors for firmness, 17 for TA, and 20 for TSS. The results of MLR did not substantially affect the prediction performance as compared to PLS. An original approach with combined chemometric and HSI data analyses was applied using R programming. Significant correlations were found between HSI data and firmness (R 2 = 0.81 and RMSE = 2.83 N) followed by TA (R 2 = 0.81 and RMSE = 0.24%) and TSS (R 2 = 0.5 and RMSE = 2.0%). Prediction maps of physicochemical qualities were achieved by applying the prediction models to each pixel of HSI to visualise their spatial distribution. The variation of firmness, TSS, and TA within the fruit indicated fruit ripening started from shoulder toward to tip part. From these results, HSI can be used as a non-destructive technique for determining the quality of fruits which could potentially enhance grading capabilities in the industrial handling and processing of mango. [ABSTRACT FROM AUTHOR]

Commercially available hyperspectral drill core scanning systems in the Visible and Near-Infrared (VNIR), Short-Wave Infrared (SWIR) and Long-Wave Infrared (LWIR) regions, along with a handheld spectrometer for the Middle-Wave Infrared (MWIR) region, were used to characterise a rank suite of coal samples (0.45% to 1.76% Rr) with different preparations (slabbed surface, outer core and compressed powder pellets). The samples used in this study came from Australian basins of Permian and Jurassic ages. The spectral characteristics of the coals measured using these scanning systems were compared to those obtained using well-established spectroscopy techniques that operate in the same region of the electromagnetic spectrum, such as Attenuated Total Reflectance Fourier Transform Infrared (ATR-FTIR) spectroscopy. The observed spectral features of the coals changed with increasing coalification. The features occurring at 1910 nm in the SWIR region and 2940 nm in the MWIR region tended to disappear as rank increased; these features arise from OH bonds related to structural water and hence correlate with the moisture content. The CH stretching fundamentals at 3280 nm correspond to the aromatic fraction of the coal, while the spectral region between 3380 nm and 3500 nm corresponds to the CH 2 and CH 3 stretching fundamentals occurring in the aliphatic fraction of the coal. The ratio of the area of the aromatic absorption feature to that of the aliphatic absorption features, herein called Rank Spectral Index, displayed a strong linear correlation with coal rank parameters, such as vitrinite reflectance. Other spectral features in the SWIR and MWIR regions also correlated with coal rank parameters. In addition to rank, trends in the magnitudes of the spectral features were influenced by maceral composition, particularly in low rank samples with abundant liptinite or inertinite group macerals. The variable spectral features related to composition were mapped onto the slabbed core surface, discriminating bright and dull bands and fusain lenses enabling the classification of coal lithotypes, regardless of rank. Among the less banded lithotypes, it was established that high liptinite content coals can be distinguished by their aliphatic spectral features in core samples. Overall, this study has shown that hyperspectral drill core scanners can be used to characterise coal and non-destructively classify rank and lithotype. • Non-destructive hyperspectral scanning systems were applied to coal characterisation. • Wavelengths covered 450 nm to 14,500 nm or visible near to thermal infrared. • Rank related features structural water, aliphatic and aromatic features were identified in the spectra. • A Rank Spectral Index was created from spectral features in the Middle-Wave Infrared. • High resolution spectral maps distinguished banding and coal lithotypes. [ABSTRACT FROM AUTHOR]

Aims: Asymptomatic nocturnal long ventricular pauses are usually detected accidentally and it has been suggested that they may lead to sudden death. Identification of predisposing factors could prevent cardiovascular events.Methods and Results: We report the case of a patient with frequent asymptomatic nocturnal ventricular pauses of 3-11 s, characteristic of a vagally mediated atrioventricular (AV) block. Echocardiography, treadmill test, thyroid function test levels, and polysomnogram were normal. In an attempt to reduce the risk, it was decided that an atrial vagal denervation induced by radiofrequency (RF) ablation (cardioneuroablation) could be useful. Spectral mapping was used to localize endocardial vagal innervation in the right and left aspects of the inter-atrial septum, responsible for the sinus node and AV node modulation, and RF pulses were applied in those sites only. After finishing the procedure, significant changes were observed in the heart rate (66-90 b.p.m.), atrial-His interval (115-74 ms), Wenckebach cycle length (820-570 ms), and sinus node recovery time (1100-760 ms). Follow-up Holter recording demonstrated that the number of ventricular pauses had reduced from 438 to 0. Heart rate and time domain characteristics were compatible with vagal denervation.Conclusion: Ablation of the endocardial vagal innervation sites seems to be safe and efficient in reducing the frequency and the length of the ventricular pauses. It was possible by identifying certain spectral components of the atrial electrogram, resulting in a conservative approach. [ABSTRACT FROM AUTHOR]

Three types of morphological features observed in different lunar crustal terrains were studied and mapped using hyperspectral Moon mineralogy mapper (M 3 ) data onboard Chandrayaan 1 mission in order to assess the utility of cascaded MGM-SAM spectral mixture modeling approach to characterize the surface materials, which may occur as mineral mixtures, at different topography of the lunar surface. Selected morphological features include: the impact melts in Orientale basin, sinuous rilles in Procellarum KREEP Terrane (PKT) and a rayed crater in Feldspathic Highland Terrane (FHT). Methodology involves extraction of spectrally pure pixels (endmembers) of the area using Pixel Purity Index (PPI), identification of mineralogy of the selected endmember spectrum using the Modified Gaussian Method (MGM) and mapping of mineralogically identified endmembers using the Spectral Angle Mapper (SAM) method. Mapping results demonstrate both the capabilities and the limitations of the MGM method of spectral deconvolution and the SAM method of spectral matching as effective tools for compositional characterizations of morphological features on the lunar surface. As a method of spectral deconvolution, MGM was able to identify and characterize both high- and low – Ca pyroxenes along with plagioclase feldspar. The Spectral Angle Mapper (SAM) was able to map identified mineral mixtures from MGM. [ABSTRACT FROM AUTHOR]

Remote sensing has been widely used to map geological structure, regolith materials and hydrothermal alteration minerals in different regions of the world, based largely on the spectral characteristics of clay/carbonate and iron oxide minerals. Given that features related to minerals may form very subtle patterns in Landsat imagery, finding suitable methods for enhancing the spectral information due to alteration minerals has been the subject of much research since the satellites were launched. Although multispectral sensors such as Landsat do not have enough spectral resolution to distinguish between specific minerals, effective processing of the data does yield useful images for regional exploration and targeting when combined with a good understanding of the associated landforms. ASTER and various airborne systems can now provide useful, high-resolution mineral information. However, the cost and time taken to acquire and process good resolution spectral data should now be balanced against the enhanced spectral discrimination possible using temporal merging, or data stacking, from the existing Landsat data archive dating back to 1972. One critical characteristic of the spectra related to surface mineral composition is its persistence over time. This study proposes a further enhancement to the data using temporal merging to create images that are both easier to interpret, and more reliable indicators of lithology and alteration. The new approach is to select a large number of scenes, typically separated by two or more years, and to merge the data for each band to give a cumulative spectral signature. The technique of stacking Landsat data to create a new dataset with much greater spectral depth is termed Landsat TM3, or Landsat TM Temporal Merge for Terrain Mapping. This emphasises not only the process used to create the dataset, but also the dominant use to which these data are ideally suited, namely mapping geological and mineral-related terrains. [ABSTRACT FROM AUTHOR]

We attempt to extract a homological structure of two kinds of graphs by the Grover walk. The first one consists of a cycle and two semi-infinite lines, and the second one is assembled by a periodic embedding of the cycles in $$\mathbb {Z}$$ . We show that both of them have essentially the same eigenvalues induced by the existence of cycles in the infinite graphs. The eigenspace of the homological structure appears as so called localization in the Grover walks, in which the walk is partially trapped by the homological structure. On the other hand, the difference of the absolutely continuous part of spectrum between them provides different behaviors. We characterize the behaviors by the density functions in the weak convergence theorem: The first one is the delta measure at the bottom, while the second one is expressed by two kinds of continuous functions, which have different finite supports $$(-1/\sqrt{10},1/\sqrt{10})$$ and $$(-2/7,2/7)$$ , respectively. [ABSTRACT FROM AUTHOR]

There are two types of voice conversion methods: statistical and deep learning-based. Although statistical methods can train with limited data, they face challenges, including spectral oversmoothing and time-domain discontinuity. On the other hand, extensively researched deep learning-based methods rely primarily on massive amounts of data, which limits their practical applicability. Given that voice conversion is an engineering problem with limited training data, it is crucial to develop techniques that can produce satisfactory results in terms of quality and similarity in the absence of a large amount of data. This paper proposes a voice conversion model based on stochastic variational deep kernel learning (SVDKL), which works with limited training data. The model allows the use of both the deep neural network's expressive capability and the high flexibility of the Gaussian process, which is a Bayesian and non-parametric method. The model utilizes a cascade of a deep neural network and a conventional kernel as the covariance function, which enables it to estimate non-smooth and more complex functions. Furthermore, the model's sparse variational Gaussian process solves the scalability problem of exact inference and enables the learning of a global mapping function for the entire acoustic space. One of the most important aspects of the proposed scheme is that the model parameters are trained using marginal likelihood optimization, which takes into account both data fitting and model complexity. Considering model complexity reduces the training data by increasing the robustness to overfitting. To evaluate the proposed scheme, we examined the model's performance with as little as approximately 80 s of training data. The results indicated that our method obtains a higher mean opinion score, smaller spectral distortion, and better preference tests than the state-of-the-art limited data methods. [ABSTRACT FROM AUTHOR]

In the present study, we have attempted the delineation of limestone using different spectral mapping algorithms in ASTER data. Each spectral mapping algorithm derives limestone exposure map independently. Although these spectral maps are broadly similar to each other, they are also different at places in terms of spatial disposition of limestone pixels. Therefore, an attempt is made to integrate the results of these spectral maps to derive an integrated map using minimum noise fraction (MNF) method. The first MNF image is the result of two cascaded principal component methods suitable for preserving complementary information derived from each spectral map. While implementing MNF, noise or non-coherent pixels occurring within a homogeneous patch of limestone are removed first using shift difference method, before attempting principal component analysis on input spectral maps for deriving composite spectral map of limestone exposures. The limestone exposure map is further validated based on spectral data and ancillary geological data. [ABSTRACT FROM AUTHOR]

Abstract: We propose a method for diagonalizing matrices with entries in commutative rings. The point of departure is to split the characteristic polynomial of the matrix over a (universal) splitting algebra, and to use the resulting universal roots to construct eigenvectors of the matrix. A crucial point is to determine when the determinant of the eigenvector matrix, that is the matrix whose columns are the eigenvectors, is regular in the splitting algebra. We show that this holds when the matrix is generic, that is, the entries are algebraically independent over the base ring. It would have been desirable to have an explicit formula for the determinant in the generic case. However, we have to settle for such a formula in a special case that is general enough for proving regularity in the general case. We illustrate the uses of our results by proving the Spectral Mapping Theorem, and by generalizing a fundamental result from classical invariant theory. [Copyright &y& Elsevier]

The high beam current and subangstrom resolution of aberration-corrected scanning transmission electron microscopes has enabled electron energy loss spectroscopy (EELS) mapping with atomic resolution. These spectral maps are often dose limited and spatially oversampled, leading to low counts/channel and are thus highly sensitive to errors in background estimation. However, by taking advantage of redundancy in the dataset map, one can improve background estimation and increase chemical sensitivity. We consider two such approaches—linear combination of power laws and local background averaging—that reduce background error and improve signal extraction. Principal component analysis (PCA) can also be used to analyze spectrum images, but the poor peak-to-background ratio in EELS can lead to serious artifacts if raw EELS data are PCA filtered. We identify common artifacts and discuss alternative approaches. These algorithms are implemented within the Cornell Spectrum Imager, an open source software package for spectroscopic analysis. [ABSTRACT FROM AUTHOR]

Abstract: We investigate simultaneous solutions of the matrix Sylvester equations , where and are k-tuples of commuting matrices of order and , respectively. We show that the matrix Sylvester equations have a unique solution X for every compatible k-tuple of matrices if and only if the joint spectra and are disjoint. We discuss the connection between the simultaneous solutions of Sylvester equations and related questions about idempotent matrices separating disjoint subsets of the joint spectrum, spectral mapping for the differences of commuting k-tuples, and a characterization of the joint spectrum via simultaneous solutions of systems of linear equations. [Copyright &y& Elsevier]

In this paper, we use artificial neural networks (ANNs) for voice conversion and exploit the mapping abilities of an ANN model to perform mapping of spectral features of a source speaker to that of a target speaker. A comparative study of voice conversion using an ANN model and the state-of-the-art Gaussian mixture model (GMM) is conducted. The results of voice conversion, evaluated using subjective and objective measures, confirm that an ANN-based VC system performs as good as that of a GMM-based VC system, and the quality of the transformed speech is intelligible and possesses the characteristics of a target speaker. In this paper, we also address the issue of dependency of voice conversion techniques on parallel data between the source and the target speakers. While there have been efforts to use nonparallel data and speaker adaptation techniques, it is important to investigate techniques which capture speaker-specific characteristics of a target speaker, and avoid any need for source speaker's data either for training or for adaptation. In this paper, we propose a voice conversion approach using an ANN model to capture speaker-specific characteristics of a target speaker and demonstrate that such a voice conversion approach can perform monolingual as well as cross-lingual voice conversion of an arbitrary source speaker. [ABSTRACT FROM AUTHOR]

We consider two fundamental properties in the analysis of two-way tables of positive data: the principle of distributional equivalence, one of the cornerstones of correspondence analysis of contingency tables, and the principle of subcompositional coherence, which forms the basis of compositional data analysis. For an analysis to be subcompositionally coherent, it suffices to analyze the ratios of the data values. A common approach to dimension reduction in compositional data analysis is to perform principal component analysis on the logarithms of ratios, but this method does not obey the principle of distributional equivalence. We show that by introducing weights for the rows and columns, the method achieves this desirable property and can be applied to a wider class of methods. This weighted log-ratio analysis is theoretically equivalent to “spectral mapping”, a multivariate method developed almost 30 years ago for displaying ratio-scale data from biological activity spectra. The close relationship between spectral mapping and correspondence analysis is also explained, as well as their connection with association modeling. The weighted log-ratio methodology is used here to visualize frequency data in linguistics and chemical compositional data in archeology. [ABSTRACT FROM AUTHOR]

Abstract: Tear fluid is a complex aqueous solution containing proteins, metabolites, electrolytes and lipids. This study uses Raman spectroscopy to analyse the composition of human tear fluid from three healthy volunteers. Two different methods are used to obtain Raman spectra from the 3μL tear samples: (i) solution-phase Raman spectroscopy, and (ii) drop coating deposition Raman spectroscopy (DCDRS). Tear samples were either basal fluid, or yawn reflex secreted fluid. Calibration of the solution technique with standard protein solutions (5–15mgmL−1) showed that this method could predict the protein concentration (cross-validation) with an error of less than 1mgmL−1. The Raman signals from the tear fluid were very weak but signals due to protein and urea were clearly observable in all samples. The drop coating deposition technique was shown to produce very high signal-to-noise spectra for relatively short acquisition times, and small sample volumes. Raman point mapping combined with principal components analysis showed that the protein, urea, bicarbonate and lipid could all be detected in the tear samples and that the distribution of these components was inhomogeneous. Their position within the drying pattern was shown to depend on their relative solubilities. The results of this study suggest that solution Raman measurements may be calibrated to give the total tear protein concentration and DCDRS could be used to give a fingerprint of the tear protein (and lipid) composition. [Copyright &y& Elsevier]

Abstract: Identification of (overlapping) communities/clusters in a complex network is a general problem in data mining of network data sets. In this paper, we devise a novel algorithm to identify overlapping communities in complex networks by the combination of a new modularity function based on generalizing NG''s Q function, an approximation mapping of network nodes into Euclidean space and fuzzy c-means clustering. Experimental results indicate that the new algorithm is efficient at detecting both good clusterings and the appropriate number of clusters. [Copyright &y& Elsevier]

Abstract: We have built two algorithms especially designed for the mapping of the intra-cluster medium emissivity and temperature, as seen by the XMM-Newton satellite using the European Photon Imaging Camera (EPIC). In both algorithms the searched parameter is firstly estimated within square resolutions elements at different spatial resolutions, then its variations are coded using Haar wavelet coefficients. The de-noised map of the searched parameter is finally obtained from structures selected as significant in the scale-space, while non-significant contributions are lowered using a regularisation process. Gas emissivity and temperature maps obtained using these algorithms have evidenced perturbations, anisotropies, shocked and disturbed regions within the two merging clusters A1750 and A3921, and have confirmed that the different component of each cluster were interacting with each other (see also Belsole, E., J-L. Sauvageot, Pratt G.W., Bourdin, H. Merging clusters of galaxies observed with XMM-Newton. Adv. Space Res., 2005 (this conference) (astro-ph/0501377)). [Copyright &y& Elsevier]

Abstract: Spectral mapping is a factorial method of multivariate data analysis that was developed around 1975 in the pharmacological research laboratory of Janssen Pharmaceutica. It has been applied in diverse fields of biology and economics where tabulated data are obtained which are defined on ratio scales. The approach has provided a few instances in which useful information was obtained that would be hard to get from bivariate analysis or from “eyeballing” of the raw data. Basically, the approach consists of logarithmic transformation and double centering of the data, factorization of the resulting cross-products matrix and a biplot of the resulting factor scores and loadings. Weighting by the marginal row and column sums of the original data can be applied in order to reduce the leverage of less important row and column items. This paper describes how the approach has fared over the past 30 years from a personal perspective. [Copyright &y& Elsevier]

Background By studying the spectrum of atrial potentials by fast Fourier transform (FFT) we have found two types of atrial muscle: the compact (CM) and the fibrillar (FM) myocardium. The former presents normal in-phase conduction inferring a great number of cellular connections, long-lasting refractoriness and leftward FFT-shift. The latter shows anisotropic out-of-phase conduction, fewer cellular connections, short refractoriness and a segmented right-FFT-shift. The compact is the normal predominant muscle and the fibrillar is different and may be neural input, vein insertion, interatrial (1A) septum, left atrial (LA) roof, etc. or pathological tissue, being so by loss of cellular connections this is a possible mechanism for conversion of compact into fibrillar-like myocardium. During atrial fibrillation (AF), clusters of FM (AF nests) present higher frequencies than any surrounding tissue. Purpose The purpose was to describe a new method for paroxysmal AF RF-ablation targeting AF nests. Method Forty patients, six control and 34 having idiopathic drug-refractory paroxysmal or persistent AF were studied and treated. Two catheters were placed in the LA by transseptal approach. RF (30–40 J/60–70 °C) was applied to all sites outside the pulmonary veins (PV) presenting right-FFT-shift (AF nests). Results Numerous AF nests were found in 34/34 AF patients and only in 1/6 controls (only in this case it was possible to induce AF despite an absence of AF history). The main FM sites were: LA roof, LA septum, close to the insertion of the superior PV, near the insertion of the inferior PV, LA posterior wall, RA near the superior vena cava insertion, RA lateral and anterior wall and the right IA septum. Ablation of all AF nests near PV insertions resulted in 35 PV isolations. After 9.9 ± 5 months only two AF patients presented relapse of a different AF form (coarse AF) which was very well controlled with medication previously ineffective. The AF was more frequent as the ratio FM/CM increased. Conclusions The RF-ablation of AF nests decreasing the fibrillar/compact myocardium ratio eliminated 94% of the paroxysmal AF in patients in the FU of 9.9 ± 5 months. The AF nests may be easily identified by spectral analysis and seem to be the real AF substrate. Paroxysmal AF may be cured or controlled by applying RF in several places outside the PV and, thereby, avoiding PV stenosis. [ABSTRACT FROM PUBLISHER]

The growth inhibitory effect of 30 synthetic dyes on 22 bacteria (test organisms) belonging to various taxonomic groups was determined. The strength (potency) and selectivity of the biological effect were separated by the spectral mapping technique, reducing the dimensionality of the selectivity maps to two by the nonlinear mapping technique. The relationship between biological effect and physicochemical parameters of dyes was elucidated by stepwise regression analysis. It has been established that the strength of the effect of anthracene and trityl derivatives was higher than that of azobenzene dyes and significantly depended on the hydrophobicity of the compound. The selectivity of the effect also depended on hydrophobicity and on the nonpolar unsaturated surface area of the dyes. Gram negative and Gram positive bacteria differed in the strength and selectivity of their response to dyes indicating the marked impact of the taxonomical position on the response. Contrary to other multivariate mathematical statistical methods biological activity may be divided by SPM into potency and selectivity values, therefore, application of the technique in future QSAR studies is highly recommended. [Copyright &y& Elsevier]

Fusion of the high-spatial-resolution hyperspectral (HHS) image using low-spatial- resolution hyperspectral (LHS) and high-spatial-resolution multispectral (HMS) image is usually formulated as a spatial super-resolution problem of LHS image with the help of an HMS image, and that may result in the loss of detailed structural information. Facing the above problem, the fusion of HMS with LHS image is formulated as a nonlinear spectral mapping from an HMS to HHS image with the help of an LHS image, and a novel cluster-based fusion method using multi-branch BP neural networks (named CF-BPNNs) is proposed, to ensure a more reasonable spectral mapping for each cluster. In the training stage, considering the intrinsic characteristics that the spectra are more similar within each cluster than that between clusters and so do the corresponding spectral mapping, an unsupervised clustering is used to divide the spectra of the down-sampled HMS image (marked as LMS) into several clusters according to spectral correlation. Then, the spectrum-pairs from the clustered LMS image and the corresponding LHS image are used to train multi-branch BP neural networks (BPNNs), to establish the nonlinear spectral mapping for each cluster. In the fusion stage, a supervised clustering is used to group the spectra of HMS image into the clusters determined during the training stage, and the final HHS image is reconstructed from the clustered HMS image using the trained multi-branch BPNNs accordingly. Comparison results with the related state-of-the-art methods demonstrate that our proposed method achieves a better fusion quality both in spatial and spectral domains. [ABSTRACT FROM AUTHOR]