Spectral Vectors Research Articles

Phase Retrieval by Linear Algebra

The null vector method, based on a simple linear algebraic concept, is proposed as an initialization method for nonconvex approaches to the phase retrieval problem. For the stylized measurement with random complex Gaussian matrices, a nonasymptotic error bound is derived, stronger than that of the spectral vector method. Numerical experiments show that the null vector method also has a superior performance for the realistic measurement of coded diffraction patterns in coherent diffractive imaging.

An advanced MRI and MRSI data fusion scheme for enhancing unsupervised brain tumor differentiation

Proton Magnetic Resonance Spectroscopic Imaging (1H MRSI) has shown great potential in tumor diagnosis since it provides localized biochemical information discriminating different tissue types, though it typically has low spatial resolution. Magnetic Resonance Imaging (MRI) is widely used in tumor diagnosis as an in vivo tool due to its high resolution and excellent soft tissue discrimination. This paper presents an advanced data fusion scheme for brain tumor diagnosis using both MRSI and MRI data to improve the tumor differentiation accuracy of MRSI alone. Non-negative Matrix Factorization (NMF) of the spectral feature vectors from MRSI data and the image fusion with MRI based on wavelet analysis are implemented jointly. Hence, it takes advantage of the biochemical tissue discrimination of MRSI as well as the high resolution of MRI. The feasibility of the proposed frame work is validated by comparing with the expert delineations, giving mean correlation coefficients for the tumor source of 0.97 and the Dice score of tumor region overlap of 0.90. These results compare favorably against those obtained with a previously proposed NMF method where MRSI and MRI are integrated by stacking the MRSI and MRI features.

Comparison of t -test ranking with PCA and SEPCOR feature selection for wake and stage 1 sleep pattern recognition in multichannel electroencephalograms

Abstract Feature selection is critical for effective analysis of data and resource savings. In multi-dimensional datasets, feature selection methods mainly use filter based approach to obtain an optimal feature subspace and wrapper methods to search for an optimal feature subset within this space. In the proposed study, two filter based statistical feature selection methods viz., statistical t -test ranking with principal component analysis (PCA) and Separability & Correlation (SEPCOR) analysis are applied to identify patterns with high discrimination between wake and stage 1 sleep of a 8-channel (6 active +2 reference electrodes) electroencephalogram (EEG) sleep dataset. The feature set consists of 6-dimensional Spectral Entropy vectors computed over EEG epochs of one second duration. In the first method, spectral entropy feature ranking is based on a t -test statistic that maximizes class separation between wakefulness/stage1 sleep. Prior to classification, PCA is performed on the ranked and non-ranked feature subsets to study the contribution of ranked channels on classifier performance. The second method uses SEPCOR analysis to automatically select an optimal feature subset with low correlation among the chosen features and maximum separation between their class means. A correlation threshold is chosen heuristically in steps of 0.05 from 0.6 to 0.75 in order to select different subsets of features. The optimal feature subsets are evaluated using multi layered perceptron (MLP) network & k-nearest neighbor (k-NN) classifiers with 50% hold out cross validation. For ranked feature subsets N = 3, 4, 5, k-NN classifier outperforms MLP network with an increase in the number of principal components (pcs). Results indicate that the pcs of ranked channels enhance the performance of k-NN classifier whereas MLP network shows only a marginal improvement with ranking for number of channels, N ≤ 4. As the number of pcs is varied from 2 to 4 in steps of one, there is an improvement of approximately 2% in the classification accuracies of k-NN classifier with ranking as compared to their non-ranked counterparts. The MLP exhibits only 1% improvement with ranking for the same case with number of hidden neurons, N = 50. The k-NN classifier responds with maximum accuracies of 96.43%, 95.7% and 94.10% (pc = 4, 3, 2 for no. of ranked channels, N = 4) as compared to 94.71%, 93.13% and 92% (pc = 4, 3, 2 non-ranked N = 4) respectively. The SEPCOR results show that with correlation threshold increasing from 0.6 to 0.75 in steps of 0.05, it automatically selects feature subsets of 2, 3, 4 and 5 which contribute to detection accuracies of 72.4%, 80%, 91.6% and 92% with k-NN classifier and improved accuracies of 73%, 85%, 95.6% and 95.8% with MLP network (no. of hidden neurons, N = 50) respectively. The SE feature ranking provides better classification results using k-NN classifier than non-ranked cases whereas features obtained using SEPCOR analysis prove to be better discriminators with MLP network for the classification of wake/stage1 sleep data. The computation speed is faster in k-NN classifier and independent of increase in value of k whereas MLP takes much more computation time for training based on the number of hidden neurons.

Implementation of Frequency Domain Approach Using Instantaneous Mixing Auto Recursive for Separation of Speech Signals

In the present work a novel algorithmic rule by taking the speech from two different microphones and separate these speeches by prediction of separating speech mixtures that is predicated on separation matrices is planned. In multi-talker applications so as to boost individual speech sources from their mixtures is done by Blind source Separation (BSS) ways. From the previous published works of separation of speech signals, the main disadvantage is that the incidence of distortion present within the signal that affects separated signal with loud musical noise. The idea for speech separation in standard BSS ways is simply one sound source in a single room. The proposed methodology uses as a network that has the parameters of the IMAR model for the separation matrices over the complete frequency vary. An attempt has been made to estimate the best values of the IMAR model parameters, ΦW and ΦG by suggests that of the maximum-likelihood estimation methodology. Based on the values of these parameters, the source spectral part vectors are estimated. The entire set of TIMIT corpus is employed for speech materials in evolution results. The Signal to Interference magnitude Relation (SIR) improves by a median of 6 dB sound unit over a frequency domain BSS approach.

Likelihood Inference for Multivariate Extreme Value Distributions Whose Spectral Vectors have known Conditional Distributions

AbstractMultivariate extreme value statistical analysis is concerned with observations on several variables which are thought to possess some degree of tail dependence. The main approaches to inference for multivariate extremes consist in approximating either the distribution of block component‐wise maxima or the distribution of the exceedances over a high threshold. Although the expressions of the asymptotic density functions of these distributions may be characterized, they cannot be computed in general. In this paper, we study the case where the spectral random vector of the multivariate max‐stable distribution has known conditional distributions. The asymptotic density functions of the multivariate extreme value distributions may then be written through univariate integrals that are easily computed or simulated. The asymptotic properties of two likelihood estimators are presented, and the utility of the method is examined via simulation.

DNN-Based Feature Enhancement Using DOA-Constrained ICA for Robust Speech Recognition

The performance of automatic speech recognition (ASR) system is often degraded in adverse real-world environments. In recent times, deep learning has successfully emerged as a breakthrough for acoustic modeling in ASR; accordingly, deep neural network (DNN)-based speech feature enhancement (FE) approaches have attracted much attention owing to their powerful modeling capabilities. However, DNN-based approaches are unable to achieve remarkable performance improvements for speech with severe distortion in the test environments different from training environments. In this letter, we propose a DNN-based FE method where the DNN inputs include preenhanced spectral features computed from multichannel input signals to reconstruct noise-robust features. The preenhanced spectral features are obtained by direction-of-arrival (DOA)-constrained independent component analysis (DCICA) followed by Bayesian FE using a hidden-Markov-model prior, to exploit the capabilities of efficient online target speech extraction and efficient FE with prior information for robust ASR. In addition, noise spectral features computed from DCICA are included for further improvement. Therefore, the DNN is trained to reconstruct a clean spectral feature vector, from a sequence of corrupted input feature vectors in addition to the corresponding preenhanced and noise feature vectors. Experimental results demonstrate that the proposed method significantly improves recognition performance, even in mismatched noise conditions.

Infrasound signal classification based on spectral entropy and support vector machine

The operation speed of the algorithm is the critical factor in the real-time monitoring of infrasound signals. The existing methods mainly focus on how to improve the accuracy of classification and can’t be used in real time monitoring because of their slow running speed. We adopt spectral entropy into the feature extraction of infrasound signals. Combined with the support vector machine algorithm, the proposed method effectively extracted the signal features meanwhile greatly improved the operation efficiency. Experimental results show that the running speed of the proposed method is 1.0s, which is far less than 4.7s of the comparison method. So the proposed method can be applied in real-time monitoring of earthquakes, tsunamis, landslides and other infrasound events.

Dynamic spectro-polarimeter based on a modified Michelson interferometric scheme.

A simple dynamic spectro-polarimeter based on a modified Michelson interferometric scheme is described. The proposed system can extract a spectral Stokes vector of a transmissive anisotropic object. Detail theoretical background is derived and experiments are conducted to verify the feasibility of the proposed novel snapshot spectro-polarimeter. The proposed dynamic spectro-polarimeter enables us to extract highly accurate spectral Stokes vector of any transmissive anisotropic object with a frame rate of more than 20Hz.

SPATIAL-SPECTRAL CLASSIFICATION BASED ON THE UNSUPERVISED CONVOLUTIONAL SPARSE AUTO-ENCODER FOR HYPERSPECTRAL REMOTE SENSING IMAGERY

Abstract. Current hyperspectral remote sensing imagery spatial-spectral classification methods mainly consider concatenating the spectral information vectors and spatial information vectors together. However, the combined spatial-spectral information vectors may cause information loss and concatenation deficiency for the classification task. To efficiently represent the spatial-spectral feature information around the central pixel within a neighbourhood window, the unsupervised convolutional sparse auto-encoder (UCSAE) with window-in-window selection strategy is proposed in this paper. Window-in-window selection strategy selects the sub-window spatial-spectral information for the spatial-spectral feature learning and extraction with the sparse auto-encoder (SAE). Convolution mechanism is applied after the SAE feature extraction stage with the SAE features upon the larger outer window. The UCSAE algorithm was validated by two common hyperspectral imagery (HSI) datasets – Pavia University dataset and the Kennedy Space Centre (KSC) dataset, which shows an improvement over the traditional hyperspectral spatial-spectral classification methods.

SPATIAL-SPECTRAL CLASSIFICATION BASED ON THE UNSUPERVISED CONVOLUTIONAL SPARSE AUTO-ENCODER FOR HYPERSPECTRAL REMOTE SENSING IMAGERY

Current hyperspectral remote sensing imagery spatial-spectral classification methods mainly consider concatenating the spectral information vectors and spatial information vectors together. However, the combined spatial-spectral information vectors may cause information loss and concatenation deficiency for the classification task. To efficiently represent the spatial-spectral feature information around the central pixel within a neighbourhood window, the unsupervised convolutional sparse auto-encoder (UCSAE) with window-in-window selection strategy is proposed in this paper. Window-in-window selection strategy selects the sub-window spatial-spectral information for the spatial-spectral feature learning and extraction with the sparse auto-encoder (SAE). Convolution mechanism is applied after the SAE feature extraction stage with the SAE features upon the larger outer window. The UCSAE algorithm was validated by two common hyperspectral imagery (HSI) datasets – Pavia University dataset and the Kennedy Space Centre (KSC) dataset, which shows an improvement over the traditional hyperspectral spatial-spectral classification methods.

Hyperspectral Blind Reconstruction From Random Spectral Projections

This paper proposes a blind hyperspectral reconstruction technique termed spectral compressive acquisition (SpeCA) conceived to spaceborne sensors systems which are characterized by scarce onboard computing and storage resources and by communication links with reduced bandwidth. SpeCA exploits the fact that hyperspectral vectors often belong to a low-dimensional subspace and it is blind in the sense that the subspace is learned from the measured data. SpeCA encoder is computationally very light; it just computes random projections (RPs) with the acquired spectral vectors. SpeCA decoder solves a form of blind reconstruction from RPs whose complexity, although higher than that of the encoder, is very light in the sense that it requires only the modest resources to be implemented in real time. SpeCA coding/decoding scheme achieves perfect reconstruction in noise-free hyperspectral images (HSIs) and is very competitive in noisy data. The effectiveness of the proposed methodology is illustrated in both synthetic and real scenarios.

Enhancing Hyperspectral Endmember Extraction Using Clustering and Oversegmentation-Based Preprocessing

Spectral mixture analysis (SMA) is an effective tool in recognition of unique spectral signatures of materials called endmembers and estimating their percentage of existence (abundance fractions). Most approaches designed in endmember extraction process are established by applying the spectral information of the dataset and, thus, tend to neglect the existing spatial correlation between adjacent pixels. Although several preprocessing modules have been developed by incorporating both spatial and spectral properties prior to spectral-based endmember extraction algorithms (EEs), they still encounter several challenges. Hence, in this paper, we propose an appropriate clustering and oversegmentation-based preprocessing (COPP) by greatly benefiting from the integration of spatial and spectral information. Moreover, a novel top-down oversegmentation (TDOS) algorithm is developed which can recognize small oversegments with high spatial correlation. Our scheme removes oversegments located at spatial border of cluster regions. Average spectral vectors of determined spatially homogenous oversegments are considered so that their spectral purity scores are calculated. COPP identifies spatially homogenous zones with the greatest spectral purity scores. Pixels of these regions are more likely to be adopted as endmembers by means of subsequent EEs. COPP can take advantage of degrading local spectral variability and noise power. The main contribution of this paper is the enhanced computational performance of EE as well as the precise reconstruction of the original hyperspectral scene besides its appropriate recognition of endmembers’ spectral signatures. The effectiveness of our design and its validation are appraised with the state-of-the-art strategies on a synthetic and AVIRIS real hyperspectral datasets.

Issues and Limitations of HMM in Speech Processing: A Survey

is the most natural way of communication among humans. This mode of communication is constituted of two parts, namely sound and sense. The intelligent production and synthesis of speech has intrigued man himself for long and efforts at automated speech recognition, has gone through various phases. Hidden Markov Models (HMMs) provide a simple and effective framework for modeling time-varying spectral vector sequences. Application of HMMs to speech recognition has seen considerable success and gained much popularity. As a consequence, almost all present day speech recognition systems are based on HMMs.

Robust snapshot interferometric spectropolarimetry

This Letter describes a Stokes vector measurement method based on a snapshot interferometric common-path spectropolarimeter. The proposed scheme, which employs an interferometric polarization-modulation module, can extract the spectral polarimetric parameters Ψ(k) and Δ(k) of a transmissive anisotropic object by which an accurate Stokes vector can be calculated in the spectral domain. It is inherently strongly robust to the object 3D pose variation, since it is designed distinctly so that the measured object can be placed outside of the interferometric module. Experiments are conducted to verify the feasibility of the proposed system. The proposed snapshot scheme enables us to extract the spectral Stokes vector of a transmissive anisotropic object within tens of msec with high accuracy.

Strategies Combining Spectral Angle Mapper and Change Vector Analysis to Unsupervised Change Detection in Multispectral Images

Change vector analysis (CVA) has been widely used in change detection of multitemporal multispectral images, whereas the use of a spectral angle mapper (SAM) is always ignored. CVA and SAM describe the difference from the perspective of vector magnitude and direction, respectively. Combining CVA and SAM can fully utilize the spectral vector and obtain a better change map. In this context, two strategies have been proposed: 1) a novel hybrid feature vector is constructed using a spectral angle and a change vector, which is utilized in SAM and CVA, respectively, and 2) a novel difference image is generated by using the autoadapted fusion strategy, which fuses the difference images acquired by utilizing CVA and SAM. During the fusion process, the autoadapted weight is defined by employing the entropy of the difference image. Experimental results on both simulated images and multitemporal multispectral images validate the effectiveness of the proposed methods.

Familiarity effects in EEG-based emotion recognition

Although emotion detection using electroencephalogram (EEG) data has become a highly active area of research over the last decades, little attention has been paid to stimulus familiarity, a crucial subjectivity issue. Using both our experimental data and a sophisticated database (DEAP dataset), we investigated the effects of familiarity on brain activity based on EEG signals. Focusing on familiarity studies, we allowed subjects to select the same number of familiar and unfamiliar songs; both resulting datasets demonstrated the importance of reporting self-emotion based on the assumption that the emotional state when experiencing music is subjective. We found evidence that music familiarity influences both the power spectra of brainwaves and the brain functional connectivity to a certain level. We conducted an additional experiment using music familiarity in an attempt to recognize emotional states; our empirical results suggested that the use of only songs with low familiarity levels can enhance the performance of EEG-based emotion classification systems that adopt fractal dimension or power spectral density features and support vector machine, multilayer perceptron or C4.5 classifier. This suggests that unfamiliar songs are most appropriate for the construction of an emotion recognition system.

Phase Noise Estimation in OFDM: Utilizing Its Associated Spectral Geometry

In this paper, we develop frequency-domain phase noise estimation schemes for an orthogonal frequency division multiplexing (OFDM) system that utilize the geometrical structure associated with the spectral components of the complex exponential of the phase noise process. This geometry can be expressed as a set of non-convex quadratic equality constraints that involve permutation matrices. Since the oscillator power spectral density is typically a low-pass process, we can reduce the dimensionality of the phase noise spectral vector. In particular, we propose a new dimensionality reduction model that preserves the phase noise spectral geometry when moving from smaller to larger dimensional spaces. Using this new model and building upon existing works, the phase noise estimation problem is posed as an optimization problem. We show, using the S-procedure, that the optimization problem has an optimal solution that is attained and can be solved using semidefinite programming. In order to reduce the computational complexity of the aforementioned optimization problem, we consider sub-optimal schemes that provide an estimate satisfying this geometry. Through numerical simulations, we demonstrate superior bit-error-rate performance when the proposed model used in conjunction with the phase noise spectral geometry is incorporated in the estimation process.

Joint-digital-predistortion for wireless transmitter’s I/Q imbalance and PA nonlinearities using an asymmetrical complexity-reduced Volterra series model

I/Q imbalance and nonlinearities of power amplifier (PA) are the main impairments of wireless transmitters degrading the spectral purity and error vector magnitude (EVM). In order to jointly migrate them, an asymmetrical complexity-reduced Volterra series model is proposed, namely joint-digital predistortion (joint-DPD). Our joint-DPD is inspired by the interaction between such two impairments, providing high degree of dynamic nonlinearities with large memory depths. Also, by understanding the structure of joint-DPD for OFDM signals, lower computational complexity can be achieved by pruning the redundant terms with an asymmetrical structure. The corresponding analysis offers the theoretical basis and its related complexity over different Volterra-series-based joint-DPDs. The performances are evaluated under two scenarios. Especially when a 3.4-dB back-off input power is applied, where severe PA nonlinearities are presented with the I/Q imbalance, the measured (left, right) adjacent channel power ratio is improved from (?19.9, ?19.6 dBc) to (?37.3, ?37.9 dBc), and the EVM is reduced from 22.1 to 3.4 %. For performance comparison, other Volterra-series-based DPDs such as memory polynomial, dynamic deviation reduction and generalized memory polynomial, are also extended to joint-DPD.

Biometric Person Identification System: A Multimodal Approach Employing Spectral Graph Characteristics of Hand Geometry and Palmprint

Biometric authentication systems operating in real world environments using a single modality are found to be insecure and unreliable due to numerous limitations. Multimodal biometric systems have better accuracy and reliability due to the use of multiple biometric traits to authenticate a claimed identity or perform identification. In this paper a novel method for person identification using multimodal biometrics with hand geometry and palmprint biometric traits is proposed. The geometrical information embedded in the user hand and palmprint images are brought out through the graph representations. The topological characterization of the image moments, represented as the virtual nodes of the palmprint image graph is a novel feature of this work. The user hand and palmprint images are represented as weighted undirected graphs and spectral characteristics of the graphs are extracted as features vectors. The feature vectors of the hand geometry and palmprint are fused at feature level to obtain a graph spectral feature vector to represent the person. User identification is performed by using a multiclass support vector machine (SVM) classifier. The experimental results demonstrate, an appreciable performance giving identification rate of 99.19% for multimodal biometric after feature level fusion of hand geometry and palmprint modalities. The performance is investigated by conducting the experiments separately for handgeometry, palmprint and fused feature vectors for person identification. Experimental results show that the proposed multimodal system achieves better performance than the unimodal cues, and can be used in high security applications. Further comparison show that it is better than similar other multimodal techniques.

Spectral polarimetric light-scattering by particulate media: 1. Theory of spectral Vector Radiative Transfer

Spectral polarimetric light-scattering by particulate media has recently attracted growing interests for various applications due to the production of directional broadband light sources. Here the spectral polarimetric light-scattering signatures of particulate media are simulated using a numerical model based on the spectral Vector Radiative Transfer Equation (VRTE). A microphysical analysis is conducted to understand the dependence of the light-scattering signatures upon the microphysical parameters of particles. We reveal that depolarization from multiple scattering results in remarkable spectral and directional features, which are simulated by our model over a wide spectral range from visible to near-infrared. We propose to use these features to improve the inversion of the scattering problem in the fields of remote sensing, astrophysics, material science, or biomedical.