Source Separation Research Articles

Blind source separation method based on blind compression transformation under impulsive noise

Blind source separation method based on blind compression transformation under impulsive noise

A Lightweight Deep Exclusion Unfolding Network for Single Image Reflection Removal.

Single Image Reflection Removal (SIRR) is a canonical blind source separation problem and refers to the issue of separating a reflection-contaminated image into a transmission and a reflection image. The core challenge lies in minimizing the commonalities among different sources. Existing deep learning approaches either neglect the significance of feature interactions or rely on heuristically designed architectures. In this paper, we propose a novel Deep Exclusion unfolding Network (DExNet), a lightweight, interpretable, and effective network architecture for SIRR. DExNet is principally constructed by unfolding and parameterizing a simple iterative Sparse and Auxiliary Feature Update (i-SAFU) algorithm, which is specifically designed to solve a new model-based SIRR optimization formulation incorporating a general exclusion prior. This general exclusion prior enables the unfolded SAFU module to inherently identify and penalize commonalities between the transmission and reflection features, ensuring more accurate separation. The principled design of DExNet not only enhances its interpretability but also significantly improves its performance. Comprehensive experiments on four benchmark datasets demonstrate that DExNet achieves state-of-the-art visual and quantitative results while utilizing only approximately 8% of the parameters required by leading methods.

Semantically complex audio to video generation with audio source separation

Semantically complex audio to video generation with audio source separation

Assessing the effect of environmental attitudes on waste material flow quality through agent-based modelling of citizen source separation behaviour.

Assessing the effect of environmental attitudes on waste material flow quality through agent-based modelling of citizen source separation behaviour.

Phylogeographic and genetic network assessment of COVID-19 mitigation protocols on SARS-CoV-2 transmission in university campus residences.

Phylogeographic and genetic network assessment of COVID-19 mitigation protocols on SARS-CoV-2 transmission in university campus residences.

An Anti-Mainlobe Suppression Jamming Method Based on Improved Blind Source Separation Using Variational Mode Decomposition and Wavelet Packet Decomposition

Mainlobe suppression jamming significantly degrades radar detection performance. The conventional blind source separation (BSS) algorithms often fail under high-jamming-to-signal-ratio (JSR) and low-signal-to-noise-ratio (SNR) conditions. To overcome this limitation, we propose an enhanced BSS method combining variational mode decomposition (VMD) and wavelet packet decomposition (WPD), termed VMD-WPD-JADE. The proposed approach first applies VMD-WPD for noise reduction in radar signals and then utilizes the JADE algorithm to compute the separation matrix of the denoised signals, effectively achieving blind source separation of radar echoes for interference suppression. We evaluate the method using noise-amplitude modulation and noise-frequency modulation jamming scenarios. The experimental results show that at a JSR = 50 dB and an SNR = −5 dB, our method successfully separates the target signals. Compared with the conventional blind source separation (BSS) algorithms, the proposed technique demonstrates superior robustness, achieving a 4–11% improvement in the target detection probability under noise-amplitude modulation (NAM) jamming and a 4–16% enhancement under noise-frequency modulation (NFM) jamming within a signal-to-noise ratio (SNR) range of −5 dB to 5 dB.

Cepstral‐Basis‐Decomposed Nonnegative Matrix Factorization for Speech Signal Modeling

This study presents an enhanced nonnegative matrix factorization (NMF) algorithm designed for speech signal modeling. NMF has demonstrated efficacy across various applications to musical instrument signals, including audio source separation and music transcription. Nevertheless, its application to speech signals often results in diminished performance due to inadequate modeling arising from the spectral continuity of the speech signal. Hence, we introduced a pioneering approach termed cepstral‐basis‐decomposed NMF (CBD‐NMF), which incorporates cepstrum analysis to enhance the modeling of speech signals. In the practical experiment, CBD‐NMF is not necessarily convergence‐guaranteed due to the flooring process; however, the experiment has revealed parameters that allow for stable optimization, ensuring that the cost function does not increase. By experimentally modeling Japanese vowel speech signals, we demonstrate that CBD‐NMF induces better representation, in which one basis arises for one mora in Japanese. Additionally, when modeling a word in Japanese speech signals, CBD‐NMF tends to induce a sparse representation equivalent to a sparse NMF with an extremely large weight coefficient. Our proposed framework can be applied to practical applications such as audio source separation and is expected to contribute to performance improvements when targeting speech signals. © 2025 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.

A Vision-Based Procedure with Subpixel Resolution for Motion Estimation.

Vision-based motion estimation for structural systems has attracted significant interest in recent years. As the design of robust algorithms to accurately estimate motion still represents a challenge, a multi-step framework is proposed to deal with both large and small motion amplitudes. The solution combines a stochastic search method for coarse-level measurements with a deterministic method for fine-level measurements. A population-based block matching approach, featuring adaptive search limit selection for robust estimation and a subsampled block strategy, is implemented to reduce the computational burden of integer pixel motion estimation. A Reduced-Error Gradient-based method is next adopted to achieve subpixel resolution accuracy. This hybrid Smart Block Matching with Reduced-Error Gradient (SBM-REG) approach therefore provides a powerful solution for motion estimation. By employing Complexity Pursuit, a blind source separation method for output-only modal analysis, structural mode shapes and vibration frequencies are finally extracted from video data. The method's efficiency and accuracy are assessed here against synthetic shifted patterns, a cantilever beam, and six-story laboratory tests.

A Novel Afterglow Molecular Probe for Monitoring of pH and Viscosity in Infected Wounds with Two-Dimensional Signal.

Organic afterglow materials have shown tremendous potential in the field of biomedical imaging. However, reports on small-molecule afterglow probes, particularly those with multitarget detection capabilities, remain limited. Here, we report a novel afterglow molecule probe (Hcy-Br-SO) that effectively responds to changes in pH and viscosity during wound infection, based on a two-dimensional (2D) signal. In this design, the enhancement of molecular afterglow performance was achieved through molecular engineering, and the underlying mechanism of afterglow emission was derived. Additionally, the synergistic enhancement of the afterglow intensity of Hcy-Br-SO by the increase in the pH and viscosity was confirmed. Besides, we observed that viscosity could retard the photoreaction process, thereby extending the duration of afterglow emission. Based on this phenomenon, we transformed the traditional time-dependent characteristics of afterglow into a measurable parameter for monitoring viscosity changes. It is noteworthy that the introduction of the time dimension not only facilitates the separation of signal sources but also explores the application potential of afterglow molecular probes. To the best of our knowledge, this is the first afterglow small-molecule probe that uses 2D signals (intensity and half-life) to monitor binocular targets. Furthermore, the Hcy-Br-SO probe was successfully used to distinguish between normal and infected wounds. This work may be useful to unravel the pathological mechanisms of chronic wounds and provide guidance for intervention.

Blind Signal Separation with Deep Residual Networks for Robust Synthetic Aperture Radar Signal Processing in Interference Electromagnetic Environments

With the rapid development of electronic technology, the electromagnetic interference encountered by airborne synthetic aperture radar (SAR) is no longer satisfied with a single type of interference, and it often encounters both suppressive and deceptive interference. In this manuscript, an algorithm based on blind signal separation (BSS) and deep residual learning is proposed for airborne SAR multi-electromagnetic interference suppression. Firstly, theoretical airborne SAR imaging in a multi-electromagnetic interference environment model is established, and the signal-mixed model of multi-electromagnetic interference is proposed. Then, a BSS algorithm using maximum kurtosis deconvolution and improved principal component analysis (PCA) is presented for suppressing the composite electromagnetic interference encountered by airborne SAR. Finally, in order to find the desired signal among multiple separated sources and to cope with the residual noise, a deep residual network is designed for signal recognition and denoising. This method uses a BSS algorithm with maximum kurtosis deconvolution and improved PCA to perform mixed signal separation. After performing signal separation, the original echo signal and the jamming can be obtained. To solve the separation order uncertainty and residual noise problems of the existing BSS algorithms, the deep residual network is designed to recognize airborne SAR signals after airborne SAR imaging. This algorithm has a better signal restoration degree, higher image restoration degree, and better compound interference suppression performance before and after anti-interference. Simulation and measurement results demonstrate the effectiveness of our presented algorithm.

Environmental, technical or health? The influence of information intervention on farmers' domestic waste separation intention.

Environmental, technical or health? The influence of information intervention on farmers' domestic waste separation intention.

Deflation properties in tensor-based eye blink removal algorithm

Blind source separation refers to a set of techniques designed to uncover latent (i.e. directly unobservable) structures in data. Depending on user preferences and the chosen algorithm, latent components can be estimated either simultaneously or iteratively, one at a time. The latter approach is typically performed using component deflation. However, Camacho et al. (Chemom Intell Lab Syst 208:104212, 2021) showed that deflation can introduce spurious artefacts into the data, particularly when the latent components are estimated under constraints. This study explored the theoretical properties of deflation in the context of higher-order arrays and tensor decomposition. In certain cases, the tensor latent components may represent noise and must be removed before further decomposition to accurately reveal the underlying structure of the data. Building on the ideas presented in Camacho et al. (Chemom Intell Lab Syst 208:104212, 2021), we investigated whether specific forms of deflation can generate spurious artefacts in electroencephalogram (EEG) tensor data, particularly under nonnegativity or unimodality constraints, where orthogonality may lack a natural interpretation. Our results are demonstrated using two real EEG datasets and one simulated dataset.

A multi-label deep residual shrinkage network for high-density surface electromyography decomposition in real-time

BackgroundThe swift and accurate identification of motor unit spike trains (MUSTs) from surface electromyography (sEMG) is essential for enabling real-time control in neural interfaces. However, the existing sEMG decomposition methods, including blind source separation (BSS) and deep learning, have not yet achieved satisfactory performance, due to high latency or low accuracy.MethodsThis study introduces a novel real-time high-density sEMG (HD-sEMG) decomposition algorithm named ML-DRSNet, which combines multi-label learning with a deep residual shrinkage network (DRSNet) to improve accuracy and reduce latency. ML-DRSNet was evaluated on a public sEMG dataset and the corresponding MUSTs extracted via the convolutional BSS algorithm. An improved multi-label deep convolutional neural network (ML-DCNN) was also evaluated and compared against a conventional multi-task DCNN (MT-DCNN). These networks were trained and tested on various window sizes and step sizes.ResultsWith the shortest window size (20 data points) and step size (10 data points), ML-DRSNet significantly outperformed both ML-DCNN (0.86 ± 0.18 vs. 0.71 ± 0.24, P < 0.001) and MT-DCNN (0.86 ± 0.18 vs. 0.66 ± 0.16, P < 0.001) in decomposition precision. Moreover, ML-DRSNet demonstrated a notably lower latency (15.15 ms) compared to ML-DCNN (69.36 ms) and MT-DCNN (76.96 ms), both of which demonstrated reduced latency relative to BSS-based decomposition methods.ConclusionsThe proposed ML-DRSNet and the improved ML-DCNN algorithms substantially enhance both the accuracy and real-time performance in decomposing MUSTs, establishing a technical foundation for neuro-information-driven motor intention recognition and disease assessment.

Single-microphone deep envelope separation based auditory attention decoding for competing speech and music

Objective.In this study, we introduce an end-to-end single microphone deep learning system for source separation and auditory attention decoding (AAD) in a competing speech and music setup. Deep source separation is applied directly on the envelope of the observed mixed audio signal. The resulting separated envelopes are compared to the envelope obtained from the electroencephalography (EEG) signals via deep stimulus reconstruction, where Pearson correlation is used as a loss function for training and evaluation.Approach.Deep learning models for source envelope separation and AAD are trained on target/distractor pairs from speech and music, covering four cases: speech vs. speech, speech vs. music, music vs. speech, and music vs. music. We convolve 10 different HRTFs with our audio signals to simulate the effects of head, torso and outer ear, and evaluate our model's ability to generalize. The models are trained (and evaluated) on 20 s time windows extracted from 60 s EEG trials.Main results.We achieve a target Pearson correlation and accuracy of 0.122% and 82.4% on the original dataset and an average target Pearson correlation and accuracy of 0.106% and 75.4% across the 10 HRTF variants. For the distractor, we achieve an average Pearson correlation of 0.004. Additionally, our model gives an accuracy of 82.8%, 85.8%, 79.7% and 81.5% across the four aforementioned cases for speech and music. With perfectly separated envelopes, we can achieve an accuracy of 83.0%, which is comparable to the case of source separated envelopes.Significance.We conclude that the deep learning models for source envelope separation and AAD generalize well across the set of speech and music signals and HRTFs tested in this study. We notice that source separation performs worse for a mixed music and speech signal, but the resulting AAD performance is not impacted.

Evaluation of oil saturation in shale from an NMR T1–T2 map based on density peak clustering and Gaussian mixture model

Fluid evaluation plays a vital role in reserve calculation and perforation location selection in shale oil exploration. Manual classification with fixed cutoffs and the blind source separation (BSS) method are the main approaches used for fluid typing in NMR T1–T2 maps. To overcome the subjectivity of the manual method and the dependence of BSS on a large dataset, an approach that integrates a clustering method, density peak clustering (DPC), and a spectral fitting technique, the Gaussian mixture model (GMM), is introduced in this paper to classify fluid components and evaluate oil saturation from T1–T2 maps. To select the number of fluid types automatically, traditional DPC is improved by defining a metric rnew based on local outlier degree, local density, relative distance, and threshold r*. The clustering accuracy of the r*−DPC method is 83.64%, which is much higher than the 52.73% accuracy of traditional DPC, and the r*–DPC method achieves better performance in fluid partitioning than other commonly used clustering algorithms. With the fluid centers information provided by r*–DPC, the GMM method is implemented to fit and extract the T1–T2 signatures of multiple fluids from T1–T2 maps. The DPC–GMM method was applied to core T1–T2 measurements and T1–T2 logging datasets, and the experimental results reveal that the average relative error of oil saturation is between 17.10% and 21.63%, which is nearly 15% lower than that of the BSS method. Furthermore, the DPC–GMM method requires less time and is easier to implement, which can make it an efficient and practical approach to fluid evaluation.

Comparative Performance of Blind Source Separation Techniques for Partial Discharge Detection in Electrical Substations

Comparative Performance of Blind Source Separation Techniques for Partial Discharge Detection in Electrical Substations

On the efficacy of psychological separation to address common method variance: Experimental evidence and a guiding research design framework.

Common method variance (CMV) substantially impacts how scholars conduct and review research. Several procedural and statistical remedies have been proposed to address the potential biasing effects that can result from CMV in data procured from a single source on a single occasion. Among them, temporal separation and distinct source designs have been the most popular. Psychological separation (PS) has also been proposed as a way to address CMV, by diverting respondents' attention from previously accessed memories, disrupting response consistency patterns, and improving effortful responding. The present research attempted to create efficacious PS through a cognitive interference task administered midway through a survey, thereby attenuating correlations that could be affected by CMV to varying degrees. In an initial study and a constructive replication, our results show that a PS intervention of at least 7.5-min attenuated several relationships to levels significantly lower than those in a single source on a single occasion design, but to an extent consistent with the attenuation achieved by temporal separation or distinct source designs. These findings suggest that under appropriate circumstances, PS is an effective strategy to address certain forms of CMV. We conclude by providing a decision guide for responsibly choosing a research design in light of various theoretical, methodological, and logistical considerations, as well as offering several additional PS task examples that can be deployed in future studies. (PsycInfo Database Record (c) 2025 APA, all rights reserved).

Aligning national waste management targets with local context: An LCA-based framework for greenhouse gas mitigation-Insights from a case study in China.

Aligning national waste management targets with local context: An LCA-based framework for greenhouse gas mitigation-Insights from a case study in China.

Collision Resolution in RFID Systems Using Antenna Arrays and Mix Source Separation

Collision Resolution in RFID Systems Using Antenna Arrays and Mix Source Separation

Unsupervised Abundance Matrix Reconstruction Transformer-Guided Fractional Attention Mechanism for Hyperspectral Anomaly Detection.

Hyperspectral anomaly detection (HAD), a challenging inverse problem, has found numerous scientific applications. Although extant HAD algorithms have achieved remarkable results, there are still several issues remained unresolved: 1) low spatial resolution (and spectral redundancy) in typical hyperspectral images prevents effectively distinguishing the abnormal pixels from those normal ones and 2) the reconstruction from existing residual-based frameworks would not completely remove anomaly effects, making the detection solely from the residual impractical. In this article, we propose a novel HAD method, termed transformer-guided fractional attention within the abundance domain (TGFA-AD), which substitutes raw input image with the abundance matrix obtained via blind source separation (BSS). First, the proposed abundance spatial-channel reconstruction transformer (ASCR-Former) is customized for rebuilding the abundance matrix. According to the image self-similarity, the abundance is patch-wisely encoded with class (CLS) tokens. The transformer encoders intensify the spatial and channel characteristics between tokens for reconstructing the abundance, followed by deriving the initial detection from the abundance residual matrix. Second, a novel fractional abundance attention (FAA) mechanism is proposed, where the attention weights coming from a specific linear combination of abundances are guided by the initial detection with convex $ Q$ -quadratic norm. Finally, the fractional convolution is incorporated to fuse the abundance and residual into the fractional feature for yielding the final detection result. Real data experiments quantitatively and qualitatively exhibit the state-of-the-art performance of TGFA-AD.