Density-based Clustering Research Articles

Anesthesia alters complexity of spontaneous and stimulus-related neuronal firing patterns in rat visual cortex

Complexity of neuronal firing patterns may serve as an indicator of sensory information processing across different states of consciousness. Recent studies have shown that spontaneous changes in brain states can occur during general anesthesia, which may influence neuronal complexity and the state of consciousness. In this study, we investigated how the firing patterns of cortical neurons, both at rest and during visual stimulation, are affected by spontaneously changing brain states under varying levels of anesthesia. Extracellular unit activity was measured in the primary visual cortex of unrestrained rats as the inhaled concentration of desflurane was incrementally reduced to 6%, 4%, 2%, and 0%. Using dimensionality reduction and density-based clustering on individual unit activities, we identified five distinct population states, which underwent dynamic transitions independent of the anesthetic level during both resting and stimulus conditions. One population state that occurred mainly in deep anesthesia exhibited a paradoxically increased number of active neurons and asynchronous spiking, suggesting a spontaneous reversal towards an awake-like condition. However, this was contradicted by the observation of low neuronal complexity in both spontaneous and stimulus-related spike activity, which more closely aligns with unconsciousness. Our findings reveal that transient neuronal states with distinct spiking patterns can emerge in visual cortex at constant anesthetic concentrations. The reduced complexity in states associated with deep anesthesia likely indicates a disruption of conscious sensory information processing.

An Empirical Sample of Spectra of M-type Stars with Homogeneous Atmospheric-parameter Labels

The discrepancies between theoretical and observed spectra, and the systematic differences between various spectroscopic parameter estimates, complicate the determination of atmospheric parameters of M-type stars. In this work, we present an empirical sample of 5105 M-type star spectra with homogeneous atmospheric parameter labels through stellar-label transfer and sample cleaning. We addressed systematic discrepancies in spectroscopic parameter estimates by adopting recent results for Gaia EDR3 stars as a reference standard. Then, we used a density-based spatial clustering of applications with noise to remove unreliable samples in each subgrid of parameters. To confirm the reliability of the stellar labels, a five-layer neural network was utilized, randomly partitioning the samples into training and testing sets. The standard deviations between the predicted and actual values in the testing set are 14 K for T eff, 0.06 dex for logg, and 0.05 dex for [M/H], respectively. In addition, we conducted an internal cross validation to enhance validation and obtained precisions of 11 K, 0.05 dex, and 0.05 dex for T eff, logg, and [M/H], respectively. A grid of 1365 high-signal-to-noise ratio (S/N) spectra and their labels, selected from the empirical sample, was utilized in the stellar parameter pipeline for M-type stars of the Large Sky Area Multi-Object Fiber Spectroscopic Telescope (LAMOST), producing an almost seamless Kiel distribution diagram for LAMOST DR10 and DR11 data. The atmospheric parameters for M-type stars from LAMOST DR11 show improved precision compared to the data from DR9, with improvements (for spectra with S/N higher than 10) from 118 to 67 K in T eff, 0.2 to 0.07 dex in logg, and 0.29 to 0.14 dex in [M/H].

PRODIGE – envelope to disk with NOEMA

Context. The formation of stars has been subject to extensive studies in the past decades on scales from molecular clouds to proto-planetary disks. It is still not fully understood how the surrounding material in a protostellar system, which often shows asymmetric structures with complex kinematic properties, feeds the central protostar(s) and their disk(s). Aims. We study the spatial morphology and kinematic properties of the molecular gas surrounding the IRS3A and IRS3B protostellar systems in the L1448N region located in the Perseus molecular cloud. Methods. We present 1 mm Northern Extended Millimeter Array (NOEMA) observations of the large program PROtostars & DIsks: Global Evolution (PRODIGE). We analyzed the kinematic properties of molecular lines. Because the spectral profiles are complex, the lines were fit with up to three Gaussian velocity components. The clustering algorithm called density-based spatial clustering of applications with noise (DBSCAN) was used to disentangle the velocity components in the underlying physical structure. Results. We discover an extended gas bridge (≈3000 au) surrounding both the IRS3A and IRS3B systems in six molecular line tracers (C18O, SO, DCN, H2CO, HC3N, and CH3OH). This gas bridge is oriented in the northeast-southwest direction and shows clear velocity gradients on the order of 100 km s−1 pc−1 toward the IRS3A system. We find that the observed velocity profile is consistent with analytical streamline models of gravitational infall toward IRS3A. The high-velocity C18O (2-1) emission toward IRS3A indicates a protostellar mass of ≈1.2 M⊙. Conclusions. While high angular resolution continuum data often show IRS3A and IRS3B in isolation, molecular gas observations reveal that these systems are still embedded within a large-scale mass reservoir, whose spatial morphology and velocity profiles are complex. The kinematic properties of the extended gas bridge are consistent with gravitational infall toward the protostar IRS3A.

An Empirical Study on User Role Discovery Based on Clustering Algorithms and Optimizations in Location-Based Social Network

Location-Based Social Network (LBSN) has been widely used in social lives. Role is an important concept in user’s personalized analysis. Many automatic methods such as machine learning method and social network analysis method have been used in user role discovery in LBSN, however, the effectiveness of these methods has not been comprehensively analyzed. In this paper, firstly, the effectiveness of five clustering algorithms is comprehensively analyzed, including K-means algorithm, Bi-Kmeans algorithm, DBSCAN (Density-Based Spatial Clustering Application with Noise) algorithm, OPTICS (Ordering points to identify the clustering structure) algorithm and Agglomerate algorithms. Secondly, four strategies are designed to optimize the algorithm for user role discovery, namely GBK-means algorithm, RDK-means (Range and density k-means) algorithm, Canopy-based algorithm and reinforcement learning based algorithm. Thirdly, six data sets are used to validate the effectiveness of these algorithms, and the result shows that the optimization strategies are effective.

Machine learning enabled uncertainty set for data-driven robust optimization

The way how the uncertainties are represented by sets plays a vital role in the performance of robust optimization (RO). This paper presents a novel approach leveraging machine learning (ML) techniques to construct data-driven uncertainty sets from historical uncertainty data for RO problems. The proposed method integrates Density-Based Spatial Clustering of Applications with Noise (DBSCAN), Gaussian Mixture Model (GMM), and Principle Component Analysis (PCA) systematically to eliminate the influence of uncertainty scenarios with low occurrence probability and generate a nonconvex uncertainty set that is a union of multiple basic subsets (box or ellipsoid) without sacrificing its computational tractability. In addition to presenting a comprehensive algorithm for uncertainty set development, this paper offers detailed guidelines for parameter tuning and performance analysis. By harnessing the well-established ML packages scikit-learn, a Python-based toolkit for implementing the proposed approach is also provided. Furthermore, a computationally efficient solution for a two-stage linear RO problem with the proposed data-driven uncertainty set is derived, alongside establishing a probabilistic guarantee of constraint satisfaction for out-of-sample uncertainties. Extensive numerical experiments, conducted on both synthetic and real-world datasets as well as an optimization-based control problem, are performed to demonstrate the efficacy of the proposed methodology.

Active domain adaptation in support of reliable operating mode detection for wind turbines

Wind turbines operate under different regimes depending on the contextual conditions, such as meteorology, operational constraints, or grid loads. Knowing when a turbine changes operational regime (its current operating mode) is crucial for a proper assessment of its performance, as only then can one compare it to the expected performance. However, often operators overseeing the turbines do not have access to the operating mode (OM) a turbine is in, leading to incomplete or inaccurate evaluation of its current capacity. We have devised an approach to leverage the knowledge across different wind farms aiming at reliably transferring, using domain adaptation, an OM detection model from a source turbine, for which OM labels are available, to a target turbine, for which they are not. This is achieved by an active learning process, enabling the adaptation to take place with expert supervision, while minimizing the required effort. Furthermore, an extension based on density-based clustering allows to post-process the outcome of the active learning to further characterize and discriminate between multiple non-primary modes of the turbine. Our approach is validated on a real-life dataset of onshore wind turbines, and we demonstrate that our active domain adaptation method achieves optimal OM detection performance.

Effect of statistical positional correlation on the radiative property investigation of dispersed particulate medium

Spectral radiative transfer between particles of dispersed particulate medium is prevalent in various fields, which may be affected by particle distribution when considering dependent scattering effect (DSE). Meanwhile, statistical positional correlation (SPC) describes the possible existing order of particle distribution in the spatial variation for disordered dispersed particulate medium. SPC plays a pivotal role in the transformation of radiative transfer regimes between particles. However, the existing theory lacks a precise criterion for describing SPC and fails to comprehensively consider key factors within SPC influencing radiative transfer regimes, such as mean distance, relative distance, standard deviation, cluster, and aggregation. To achieve more representative and accurate results of radiative property calculations while minimizing computational resources, we proposed MRSDL (Mean distance, Relative distance, Standard deviation, Density-based clustering, and Line matrix) criterion combined with particle swarm optimization (PSO) for characterizing and regulating SPC. Moreover, to further achieve the statistical averaging of radiative properties precisely, the multiple sphere T-matrix (MSTM) method is combined to eliminate the random fluctuations of radiative properties caused by SPC. Compared to the conventional method, the method by authors can decrease error between experimental and calculation data from 41.23 % to 5.32 %, when considering the effect of SPC on the radiative property.

Near-Earth stream decoherence revisited: the limits of orbital similarity

Orbital similarity measures, such as the D values, have been extensively used in meteor science to identify meteoroid streams and associate meteorite falls with near-Earth objects (NEOs). However, the chaotic nature of near-Earth space challenges the long-term reliability of these measures for stream identification, and the increasing size of our fireball, meteorite fall, and NEO databases make random associations more common. Despite this, many researchers erroneously continue to use orbital similarity beyond its inherent limits. We aim to assess the statistical significance of using orbital similarity measures for identifying streams of meteoroids or asteroids and explore the implications of chaotic dynamics on the long-term coherence of these streams. We employed a kernel density estimation (KDE) based method to evaluate the statistical significance of orbital similarities within different datasets. Additionally, we conducted a Lyapunov characteristic lifetime analysis and simulated 300 fictitious meteoroid streams to estimate the decoherence lifetimes in near-Earth space. The orbital similarity was determined using the D$_ SH $, D', and D$_ H $ orbital similarity discriminants. Clustering analysis relied on a density-based spatial clustering of applications with noise (DBSCAN) algorithm. Our analysis found no statistically significant streams within the meteorite fall, fireball, or USG impact datasets, with orbital similarities consistent with random associations. Conversely, 12 statistically significant clusters were identified within the NEO population, likely resulting from tidal disruption events. The Lyapunov lifetime analysis revealed short characteristic lifetimes (60–200 years) for orbits in near-Earth space, emphasizing the rapid divergence of initially similar orbits. Meteoroid stream decoherence lifetimes ranged from $10^4$ to $10^5$ years, aligning with previous studies and underscoring the transient nature of such streams. The rapid decoherence of meteoroid streams and the chaotic dynamics of near-Earth orbits suggest that no reported stream or NEO associations of meteorites or fireballs are statistically significant according to orbital discriminates. Many are likely coincidental rather than indicative of a true physical link. However, several statistically significant clusters found within the NEO population are consistent with a tidal disruption formation. This contrast and lack of statistically significant associations amongst the impact datasets is likely due to the fireball databases being 2 orders of magnitude smaller than the NEO database and the higher intrinsic uncertainties of fireball observation derived orbits.

Locating Strong Electromagnetic Pulses Recorded by a Single Satellite with Cluster Analysis and Worldwide Lightning Location Network Observations

The integration of satellite-borne and ground-based global lightning location networks offers a better perspective to study lightning processes and their evolutionary characteristics within thunderstorm clouds, thereby bolstering the predictive capabilities for severe weather phenomena. Currently, the satellite-borne network is in the preliminary testing phase with a single satellite. The geographic locations of single-satellite detection events primarily rely on synchronous information from coincident ground-based network events; this method is called synchronous locating (SCL). However, variations in detection-frequency bands and system capabilities prevent this method from accurately locating more than a mere 10% of events. To address this limitation, this paper introduces a cluster-analysis-based strategy, utilizing the observations from the Worldwide Lightning Location Network (WWLLN), termed the cluster analysis locating (CAL) method. The CAL method’s performance, influenced by the density-based spatial clustering of applications with noise (DBSCAN), the K-means, and the mean shift algorithms, is examined. Subsequently, an advanced version, mean shift denoised (MSDN)-CAL, is proposed, demonstrating marked improvements in location accuracy and reliability over the other CAL methods. The satellite-borne wideband electromagnetic pulse detector (WEMPD), orbiting at an altitude of approximately 500 km with a 97.5° inclination, captured 1061 strong electromagnetic pulses (EMPs). Among these, trans-ionospheric single pulses (TISPs) and trans-ionospheric pulse pairs (TIPPs) constituted 21.30% and 78.70%, respectively. Using the MSDN-CAL method successfully determines the geographic locations for 81.15% (861 out of 1061) of the events. This success rate represents an approximate eightfold enhancement over the SCL method. The arithmetic mean, geometric mean, and standard deviation of the two-dimensional range deviation of the locating results between the MSDN-CAL method versus the WWLLN-SCL (or the Guangdong-Hong Kong-Macao Lightning Location System (GHMLLS)-SCL) method are 51.06 (176.26) km, 16.17 (92.53) km, and 100.95 (174.79) km, respectively. Furthermore, it has been possible to estimate the occurrence altitudes for 81.92% (684 out of 835) of the TIPP events. The altitude deviations, as determined by comparing them with the GHMLLS-SCL method’s locating results, exhibit an arithmetic mean of 2.08 km, a geometric mean of 1.30 km, and a standard deviation of 2.26 km. The outcomes of this research establish a foundation for deeper investigation into the origins of various event types, their seasonal variations, and their geographical distribution patterns. Moreover, they pave the way for utilizing a single satellite to measure global surface reflectance, thus contributing valuable data for meteorological and atmospheric studies.

Development of HTC-DBSCAN: A Hierarchical Trajectory Clustering Algorithm with Automated Parameter Tuning

Existing route-clustering methods often fail to identify abnormal sections or similarities between routes, mainly when working with large or long datasets. While sub-route clustering can detect regional patterns, it struggles to accurately capture the overall route structure. The present study proposes a new ship route-clustering method that enhances computational efficiency and noise recognition while addressing these limitations. We refined Automatic Identification System data via four data-cleaning processes and applied a statistical distance measurement to assess ship trajectory similarity. Dimensionality reduction was then used to facilitate clustering. The clustering of ship route similarities is non-parametric and can be applied to datasets not separated based on density to find clusters of various densities. Density-Based Spatial Clustering of Applications (DBSCA) applies to many research fields; using the DBSCA with Noise (DBSCAN) algorithm, we propose an improved DBSCAN algorithm that automatically determines the parameters Epsilon and MinPts. In this study, as a core ship route-clustering process, we propose a sub-route clustering process by setting the distance and density of data points to clear standards for re-analysis and completion. The proposed approach demonstrates markedly enhanced clustering performance, offering a more sophisticated and efficient basis for ship route decision-making.

Root segmentation of horticultural plants in X-Ray CT images by integrating 2D instance segmentation with 3D point cloud clustering

X-ray computed tomography (CT) is a powerful tool for in situ plant root system architecture (RSA) characterization. Accurate root segmentation from CT images is integral to studying RSA. Research studies on segmenting roots from CT images have been mainly limited to image processing-based approaches which may require parameter tuning and often lack common segmentation metrics, e.g., Dice and IoU. A recent deep learning approach utilizes a volumetric encoder-decoder network to achieve a high Dice score and IoU. However, training a volumetric model is dependent on the availability of fully annotated scans of the growing medium column, obtaining which can be time-consuming, tedious, and resource intensive. In this study, an efficient method using deep learning-based instance segmentation in conjunction with density-based spatial clustering of applications with noise (DBSCAN)-based filtering was developed and evaluated for two horticultural plant species. A pretrained Mask R-CNN model was fine-tuned on images selected along different axes of the three-dimensional scans to identify the best view selection strategy for volumetric root segmentation. DBSCAN was used to filter noise from the volumetric segmentation with an automated parameter tuning technique. The proposed method was evaluated on scans of poinsettias and onions and achieved best average scores of 0.831, 0.839, 0.834, and 0.718 for Precision, Recall, Dice, and IoU, respectively. Further experiments showed reducing the training data to 1 % did not significantly impact the segmentation accuracy. Therefore, the proposed method has promising potential to facilitate RSA analysis with its high utility.

SingletSeeker: an unsupervised clustering approach for automated singlet discrimination in cytometry.

Flow cytometry is a high-throughput, high-dimensional technique that generates large sets of single-cell data. Prior to analyzing this data, it is common to exclude any events that contain two or more cells, multiplets, to ensure downstream analysis and quantification is of single-cell events, singlets, only. The process of singlet discrimination is critical yet fundamentally subjective and time-consuming; it is performed manually by the user, where the proper exclusion of multiplets depends on the user's expertise and often varies from experiment to experiment. To address this problem, we have developed an algorithm to automatically discriminate singlets from other unwanted events such as multiplets and debris. Using parameters derived from imaging, the algorithm first identifies high-density clusters of events using a density-based clustering algorithm, and then classifies the clusters based on their properties. Multiplets are discarded in the first step, while singlets are distinguished from debris in the second step. The algorithm can use different strategies on imaging feature selection-based user's preferences and imaging features available. In addition, the relative importance of singlets precision vs. sensitivity can be further tweaked via a density coefficient adjustment. Twenty-two datasets from various sites and of various cell types acquired on the BD FACSDiscover™ S8 Cell Sorter with CellView™ Image Technology were used to develop and validate the algorithm across multiple imaging feature sets. A consistent singlets precision >97% with a solid >88% sensitivity has been demonstrated with a LightLoss feature set and the default density coefficient. This work yields a high-precision, high-sensitivity algorithm capable of objective and automated singlet discrimination across multiple cell types using various imaging-derived parameters. A free FlowJo™ Software plugin implementation is available for simple and reproducible singlet discrimination for use at the beginning of any user's workflow.

A method for signal components identification in acoustic signal with non-Gaussian background noise using clustering of data in time-frequency domain

This paper presents a novel method for fault detection in vibration/acoustic signals contaminated with non-Gaussian noise, specifically addressing the challenge of random impulsive and wideband disturbances in industrial measurements. While damage detection in Gaussian noise environments is well understood, high-amplitude non-cyclic impulsive disturbances arising from random aspects of industrial processes, such as non-uniform operations and random impacts, pose significant analytical challenges.The proposed method analyzes the distribution densities of spectral vectors derived from spectrograms. It considers a simple additive model consisting of the signal of interest (SOI) and Gaussian and non-Gaussian noise. Using the density-based spatial clustering algorithm (DBSCAN), the method isolates distinct classes of spectral vectors from the spectrogram, effectively separating different signal behaviors and extracting fault-related information. The effectiveness of the proposed method was validated using an envelope spectrum-based indicator (ENVSI) and successfully demonstrated on real signals from an industrial machine with a faulty bearing.

Exploration of Halo Substructures in Integrals-of-motion Space with Gaia Data Release 3

Using kinematic data from the Gaia Data Release 3 catalog, along with metallicity estimates robustly derived from Gaia BP/RP spectra, we have explored the Galactic stellar halo in search of both known and potentially new substructures. By applying the Hierarchical Density-Based Spatial Clustering of Applications with Noise clustering algorithm in integrals-of-motion space (i.e., E, L z , and L ⊥ =Lx2+Ly2 ), we identified five previously known substructures: Gaia-Sausage-Enceladus (GSE), Helmi streams, I'itoi and Sequoia, and the hot thick disk. We additionally found NGC 3201 and NGC 5139 in this work, and NGC 3201 shares similar distributions in phase space and metallicities to Arjuna, which possibly implies that they have the same origin. Three newly discovered substructures are Prograde Substructure 1 (PG1), Prograde Substructure 2 (PG2), and the Low Energy Group. PG1, with a higher V ϕ than typical GSE member stars, is considered as either a low-eccentricity and metal-rich part of GSE or part of the metal-poor disk. PG2, sharing kinematic similarities with Aleph, is thought to be its relatively highly eccentric component or the mixture of Aleph and the disk. The Low Energy Group, whose metal-poor component of metallicity distribution function has a mean value [M/H] ∼ −1.29 (compared to that of Heracles [M/H] ∼ −1.26), may have associations with Heracles.

Unsupervised machine learning for local stress identification in fatigue analysis of welded joints

AbstractIn the underlying study, a method has been proposed to automatically extract finite element (FE) peak stresses of welded components to alleviate human errors and increase the calculation accuracy. The approach is based on the K-means and DBSCAN (density-based spatial clustering of applications with noise) methods as the unsupervised machine learning approaches. Data points, in this case, nodal coordinates and their corresponding stress magnitudes, are grouped within different clusters. The peak stress in each dense region (cluster) is then highlighted and reported automatically. Parametric and comparative studies have also been carried out in order to detect optimised parameters of the K-means and DBSCAN algorithms. The methodology will ultimately be used for more reliable stress analysis in fatigue assessment of welded structures.

Retinal blood vessel segmentation using density-based fuzzy C-means clustering and vessel neighborhood connected component

Retinal blood vessel segmentation is a crucial process in medical image analysis. However, it is often challenging due to the variation in color, shape, intensity, size, and contrast of blood vessels. Most of the relevant studies concentrate on algorithms based on supervised learning and few on deep learning. However, due to the several challenges in retinal image acquisition, these algorithms cannot deliver the highest possible level of accuracy. Therefore, this paper implements retinal blood vessel segmentation and classification (RBVSC) methods using density-based fuzzy C-means clustering and vessel neighborhood-connected components, hereafter denoted as DBFCM-VNCC. Initially, the given retinal images are preprocessed using the contrast enhancement method that involves Histogram Equalization with Variable Enhancement Degree (HEVED), selecting the appropriate color channel, optic disc elimination, and using a Gaussian filter, which removes the noise or artifacts from the retinal images. Then, a fully fuzzy C-means clustering is used for coarse segmentation of vessel lesions, which can detect the affected blood vessel features quite efficiently. Finally, structure-based algorithms based on vessel neighborhood-connected components based on mathematical dilation operators and local thicknesses are used to obtain accurate skeletonization and segmentation of the retinal vessels. The algorithm was assessed using three open-access retinal image databases: DRIVE, CHASE_DB1, and HRF, where it achieved mean sensitivity, specificity, accuracy, area overlap measure, and error rate scores of 98.16%, 98.74%, 97.68%, and 4.54%; 98.25%, 98.81%, 97.68%, 97.86%, and 2.14%; 98.22%, 98.78%, 97.56%, 97.40%, and 2.60% for segmenting the retinal vessel. This demonstrates the effectiveness of the proposed DBFCM-VNCC techniques.

M-CLUSTER: multistage clustering for unsupervised train wheel condition monitoring

Wheel out-of-roundness (OOR) is a common wheel defect that raises maintenance costs and increases the risk of failure or damage to track components. This paper proposes a novel multistage clustering framework (M-CLUSTER) for unsupervised condition monitoring of train wheels. The framework initially extracts time-domain features from raw acceleration responses collected by rail-mounted sensors. Sensitive features are then selected through an unsupervised feature selection algorithm called local learning-based clustering (LLC). Next, a detector model is trained using density-based spatial clustering of applications with noise (DBSCAN), a data clustering method effective for clusters with similar density. Since this algorithm does not originally involve separate training and testing phases, a new two-step mechanism is introduced: (1) training on a healthy dataset and (2) testing on an unlabelled dataset. Finally, the severity of train wheel defects is classified by K-means, with cluster validity indices (CVI) automatically determining the number of severity clusters (classes). The framework’s efficiency is demonstrated through the detection of defective wheels using the Alfa Pendular passenger model. Results indicate that M-CLUSTER accurately identifies train wheel flats and polygonal wear without labelled data, achieving 98% accuracy by selecting 10 features from the set.

Integrated STL-DBSCAN algorithm for online hydrological and water quality monitoring data cleaning

Online hydrological and water quality monitoring data has become increasingly crucial for water environment management such as assessment and modeling. However, online monitoring data often contains erroneous or incomplete datasets, consequently affecting its operational use. In the study, we developed an automated data cleaning algorithm grounded in Seasonal-Trend decomposition using Loess (STL) and Density-Based Spatial Clustering of Applications with Noise (DBSCAN). STL identifies and corrects more obvious anomalies in the time series, followed by DBSCAN for further refinement, in which the reverse nearest neighbor method was employed to enhance the clustering accuracy. To improve anomaly detection, a two-level residual judgment threshold was applied. The algorithm has been successfully applied to three reservoirs in Shanghai, China, achieving the precision rate of 0.91 and recall rate of 0.81 for dissolved oxygen and pH. The proposed algorithm can be potentially applied for cleaning of environment monitoring data with high accuracy and stability.

Replay-Based Incremental Learning Framework for Gesture Recognition Overcoming the Time-Varying Characteristics of sEMG Signals.

Gesture recognition techniques based on surface electromyography (sEMG) signals face instability problems caused by electrode displacement and the time-varying characteristics of the signals in cross-time applications. This study proposes an incremental learning framework based on densely connected convolutional networks (DenseNet) to capture non-synchronous data features and overcome catastrophic forgetting by constructing replay datasets that store data with different time spans and jointly participate in model training. The results show that, after multiple increments, the framework achieves an average recognition rate of 96.5% from eight subjects, which is significantly better than that of cross-day analysis. The density-based spatial clustering of applications with noise (DBSCAN) algorithm is utilized to select representative samples to update the replayed dataset, achieving a 93.7% recognition rate with fewer samples, which is better than the other three conventional sample selection methods. In addition, a comparison of full dataset training with incremental learning training demonstrates that the framework improves the recognition rate by nearly 1%, exhibits better recognition performance, significantly shortens the training time, reduces the cost of model updating and iteration, and is more suitable for practical applications. This study also investigates the use of the incremental learning of action classes, achieving an average recognition rate of 88.6%, which facilitates the supplementation of action types according to the demand, and further improves the application value of the action pattern recognition technology based on sEMG signals.

Enhanced TomoSAR imaging method with complex least squares for 3D building reconstruction

ABSTRACT Synthetic aperture radar tomography (TomoSAR) has emerged as an advanced technology for the rapid acquisition of three-dimensional information, with model solutions dependent on a specific complex least squares adjustment criterion. We design two TomoSAR imaging algorithms based on two adjustment criterions minimizing the square sum of the real and imaginary components of observation vector residual, and minimizing the square sum of the L2-norm of observation vector residual. Additionally, we propose a strong scatterer recognition method that combines amplitude thresholding and density-based spatial clustering of applications with noise to alleviate the low efficiency and excessive noise in real images. The simulated and practical experiments are conducted to evaluate the performance of algorithms. Results suggested that two algorithms perform similarly in parameter estimation accuracy and noise immunity for buildings with simple structures and single scatterer. However, for buildings with multiple scatterers, the algorithm derived by minimizing the square sum of the L2-norm from the observation vector residual exhibits more robustly than the other algorithm, accurately distinguishing multiple scatterers within pixels and reconstructing 3D scenes mirroring the actual ones. The results verify the effectiveness of the proposed strong scatterer recognition method and the importance of the adjustment criterion in TomoSAR imaging.