Independent component analysis (ICA) utilizing prior information, also called semiblind ICA, has demonstrated considerable promise in the of functional magnetic resonance imaging (fMRI). So far, temporal information about fMRI been used ICA or spatial as additional constraints to improve estimation task-related components. Considering that patterns is available, a algorithm was proposed within framework constrained with fixed-point learning. The approach first tested synthetic fMRI-like...

The Empirical Mode Decomposition (EMD) is a general signal processing method for analyzing nonlinear and nonstationary time series. central idea of EMD to decompose series into finite often small number intrinsic mode functions (IMFs). An IMF defined as any function having the extrema zero-crossings equal (or differing at most by one), also symmetric envelopes local minima, maxima respectively. decomposition procedure adaptive, data-driven, therefore, highly efficient. In this contribution,...

Convolutional neural networks (CNNs) have shown promising results in classifying individuals with mental disorders such as schizophrenia using resting-state fMRI data. However, complex-valued data is rarely used since additional phase introduces high-level noise though it potentially useful information for the context of classification. As such, we propose to use spatial source (SSP) maps derived from CNN input. The SSP are not only less noisy, but also more sensitive activation changes...

Defocus blur detection (DBD) is a fundamental yet challenging topic, since the homogeneous region obscure and transition from focused area to unfocused gradual. Recent DBD methods make progress through exploring deeper or wider networks with expense of high memory computation. In this paper, we propose novel learning strategy by breaking problem into multiple smaller defocus detectors thus estimate errors can cancel out each other. Our focus diversity enhancement via cross-ensemble network....

The underdetermined problem poses a significant challenge in blind source separation (BSS) where the number of signals is greater than that mixed signals. Motivated by fact security many cryptosystems relies on apparent intractability computational problems such as integer factorization problem, we exploit BSS to present novel BSS-based speech encryption properly constructing mixing matrix for encryption, and generating key satisfy necessary condition proposed method be unconditionally...

Joint blind source separation (J-BSS) is an emerging data-driven technique for multi-set data-fusion. In this paper, J-BSS addressed from a tensorial perspective. We show how, by using second-order statistics in J-BSS, specific double coupled canonical polyadic decomposition (DC-CPD) problem can be formulated. propose algebraic DC-CPD algorithm based on rank-1 detection mapping. This converts possibly underdetermined to set of overdetermined CPDs. The latter solved algebraically via...

We consider tensor data completion of an incomplete observation multidimensional harmonic (MH) signals. Unlike existing tensor-based techniques for MH retrieval (MHR), which mostly adopt the canonical polyadic decomposition (CPD) to model simple "one-to-one" correspondence among harmonics across difference modes, we herein use more flexible block term (BTD) that can be used describe complex mutual correspondences several groups different modes. An optimization principle aims fit BTD in least...

Non-orthogonal joint diagonalization (NJD) free of prewhitening has been widely studied in the context blind source separation (BSS) and array signal processing, etc. However, NJD is used to retrieve jointly diagonalizable structure for a single set target matrices which are mostly formulized with dataset, thus insufficient handle multiple datasets inter-set dependences, scenario often encountered BSS (J-BSS) applications. As such, we present generalized (GNJD) algorithm simultaneously...

A nanometric barium titanate (BaTiO3) film was fabricated on a silicon (SiO2/Si) substrate by the spin-coating method to make resistive-type humidity sensor. The sensing properties of sensor were measured, including sensitivity, hysteresis, and response recovery times. did not change with level applied voltage. best linearity curve impedance versus relative (RH) appears at frequency 100 Hz. capacitance increases as RH increases, especially for low measurement frequency. peaks dielectric loss...

Direction of arrival (DOA) estimation is the basis for underwater target localization and tracking using towed line array sonar devices. A method DOA wideband weak targets based on coherent signal subspace (CSS) processing compressed sensing (CS) theory proposed. Under CSS framework, frequency focusing accompanied by a two-sided correlation transformation, allowing to be estimated spatial sparsity reconstruction algorithm. Through analysis simulation data marine trial data, it shown that...

Tucker decomposition can provide an intuitive summary to understand brain function by decomposing multi-subject fMRI data into a core tensor and multiple factor matrices, was mostly used extract functional connectivity patterns across time/subjects using orthogonality constraints. However, these algorithms are unsuitable for extracting common spatial temporal subjects due distinct characteristics such as high-level noise. Motivated successful application of image denoising the intrinsic...

Canonical polyadic decomposition (CPD) of multi-subject complex-valued fMRI data can be used to provide spatially and temporally shared components among groups with both magnitude phase information. However, the CPD model is not well formulated due large subject variability in spatial temporal modalities, as high noise level complexvalued data. Considering that shift-invariant across subjects, we propose further impose a sparsity constraint on maps denoise inter-subject well. More precisely,...

Abstract Spatial source phase, the phase information of spatial maps extracted from functional magnetic resonance imaging (fMRI) data by data‐driven methods such as independent component analysis (ICA), has rarely been studied. While observed shown to convey unique brain information, role in representing intrinsic activity is yet not clear. This study explores for identifying differences between patients with schizophrenia (SZs) and healthy controls (HCs) using complex‐valued resting‐state...

Convolutional neural networks (CNN) have exhibited great success in image classification. The application of CNN to classification patients with brain disorders and healthy controls is also promising using functional magnetic resonance imaging (fMRI) data. However, the shortage number subjects a challenge for training CNN. Spatial maps separated from fMRI data by independent component analysis (ICA) can provide solution this problem within an ICA-CNN framework. As such, we propose three...

Blind source separation (BSS) is explored to add another encryption level besides the existing methods for image cryptosystems. The transmitted images are covered with a noise by specific mixing before and then recovered through BSS after decryption. Simulation results illustrate validity of proposed method.

This paper considers target localization with a multistatic MIMO radar system of multiple transmit arrays and receive arrays. We formulate the matched filtered output data into tensors that admit double coupled canonical polyadic decomposition (DC-CPD) model, which efficiently characterizes coupling between arrays, multilinear structure in each tensor. propose novel algebraic DC-CPD algorithm based on rank-1 detection mapping, provides analysis uniqueness conditions. A post-processing...

Focused beamformers are widely used in passive underwater acoustic localization. Many pseudo-peaks (spurious peaks not corresponding to a real source) produced by focused beamformers, impacting the performance and reliability of related algorithms. After describing both received signal model near-field sources with uniform line array, basic principle localization based on we define beamformer's spectra give procedure quantify these pseudo-peaks. The conventional (C-FBs) minimum variance...

Prior work has reported that brain functional networks can be utilized to differentiate healthy subjects and patients with mental disorder. Group independent component analysis (GICA) is a widely-used data-driven method for extracting from resting-state magnetic resonance imaging (fMRI) data of multiple subjects. GICA approaches estimate the group-level components first, then back-reconstruct subject-specific their associated time courses based on components. To networks, previous studies...