Search

Defect detection is crucial in ensuring the quality of steel products. This paper proposes a novel deep neural network, cascaded adaptive global location network (CAGLNet), for detecting steel surface defects. The main objective of this study is to address the challenges associated with the irregular shape and dense spatial distribution of defects on steel. To achieve this goal, CAGLNet integrates a feature extraction network that combines residual and feature pyramid networks, a cascade adaptive tree-structure region proposal network (CAT-RPN) that eliminates the need for prior knowledge, and a global localisation regression for steel defect detection. This paper evaluates the effectiveness of CAGLNet on the NEU-DET dataset and demonstrates that the proposed model achieves an average accuracy of 85.40% with a fast frames per second of 10.06, outperforming those state-of-the-art methods. These results suggest that CAGLNet has the potential to significantly improve the effectiveness of defect detection in industrial production processes, leading to increased production yield and cost savings. Abbreviations: AT-RPN, adaptive tree-structure region proposal network; CAGLNet, cascaded adaptive global location network; CAT-RPN, cascade adaptive tree-structure region proposal network; CNN, convolutional neural network; DNN, deep neural network; EPNet, edge proposal network; FPN, feature pyramid network; FCOS, fully convolutional one-stage detector; FPS, frames per second; GMM, Gaussian mixture model; IoU, intersection-over-union; ROIAlign, region of interest align; RPN, region proposal network; ResNet, residual network; ResNet50_FPN, residual network and feature pyramid network; SABL, side aware boundary localisation; SSD, single-shot multiBox detector; TPE, Tree-structured Parzen estimator [ABSTRACT FROM AUTHOR]

The competition among Food & Beverage companies has substantially increased in today's age of digitization. Sales forecasting is one of their main challenges. Due to space limitations, employee shortages, and rising online demand, retail sales forecasting became extremely important for Food and Beverage companies. This research analyzed the sales data of a multinational Food & Beverage Company. It proposed a framework using Gaussian Mixture Model (GMM) clustering, Hierarchical Agglomerative Clustering (HAC), and Random Forest algorithm for forecasting sales. This model analyzes the impact of the weekends, holidays, promotional activities, customer sentiments, festivals, and socio-economic situations in sales data and is able to forecast sales ranging from one to 15 months. An investigation of the suggested model's performance compared to numerous cutting-edge sales forecasting techniques is carried out to show its efficacy. Here, we demonstrate that the proposed hybrid model surpasses current predicting and computing efficiency methods. The results of this study can help retail managers to allocate resources and manage inventories in well-informed ways. The findings suggest that combining many strategies may produce the most precise forecasts. [ABSTRACT FROM AUTHOR]

Wire arc additive manufacturing (WAAM) is an additive manufacturing process for building large-sized metal components using gas metal arc welding technology. Detecting defects during deposition is critical for halting production of low-quality components, thereby reducing waste and associated costs, and allowing timely process adjustments. This necessitates the integration of online anomaly detection systems in modern WAAM systems. While most existing methods rely on high-frequency data acquisition, this research explores the feasibility of using low-cost and low-frequency acquisition systems for anomaly detection, leveraging the advancements in Industry 4.0 and IoT protocols like MQTT. This study presents an IoT-driven intelligent system for WAAM, where data is collected at a low frequency of 10 Hz and processed using unsupervised machine learning techniques to develop an anomaly detection service. The methodology involves using a data-driven model to forecast WAAM process variables and detect anomalies through a Gaussian mixture model based on estimation errors between model estimation and real values collected from the process. The results obtained from various models were compared, specifically polynomial autoregression with exogenous variables (polyARX), autoregressive neural network with exogenous variables (ARXNN), and long short-term memory (LSTM). The proposed system demonstrated high performance in anomaly detection, achieving an F2-score of 90.4% with the LSTM model and 86.28% with the ARXNN model when used to forecast the WAAM process. These results are comparable to performance reached with supervised machine learning algorithms or with results reached by previous studies which employ high-frequency data. Furthermore, the results show that low-frequency data, when processed with complex ML techniques, can reduce the chance of missing anomalies by 40% compared to traditional statistical methods, thanks to a higher recall. Finally, the implementation of the system using containerisation technologies like Docker was discussed, and post-deployment results confirmed the findings. Additionally, a graphical user interface was developed for operator interaction, utilising emerging IoT technologies such as Grafana and SQL databases, which represent additional parts of the IoT-driven intelligent WAAM system proposed in this work. [ABSTRACT FROM AUTHOR]

Pine wilt disease poses a serious threat to pine forests worldwide. It causes rapid pine tree mortality, profoundly impacting ecosystems and economic assets. Using unmanned aerial vehicle (UAV) imagery to identify, locate, and promptly remove dead pine trees in forest areas is an effective method to control the spread of pine wilt disease. However, dataset annotation and validation remains challenging. To address the high cost and time-consuming nature of manual annotation in fully supervised learning for identifying dead pine trees, this study introduces semi-supervised learning framework for detecting dead pine trees. Unlike traditional fully supervised learning methods, our approach reduces the need for a large quantity of precisely annotated data. By incorporating attention-enhanced pyramid and adaptive thresholding mechanisms, the model can effectively learn from a small amount of labelled data and generate high-quality pseudo-labels from unlabelled data. Specifically, the model employs the channel prior convolutional attention mechanism in feature pyramid fusion to enhance sensitivity to key features of dead trees and improve consistency learning. Meanwhile, probabilistic clustering Gaussian Mixture Model is also used to adaptively adjust the threshold for pseudo-labels assignment, enhancing the accuracy of pseudo-labels during training. Experiments were conducted using self-constructed dead pine tree image datasets to validate the effectiveness of the proposed method. The results show that our method requires only 25% of the labelled data to achieve recognition results close to those of fully supervised method with a fully labelled. Additionally, its performance surpasses existing semi-supervised methods. The proposed algorithm provides an efficient and cost-effective approach for detecting dead pine trees for pine wilt disease control using UAV imagery. [ABSTRACT FROM AUTHOR]

Transport electrification is a crucial element of the ongoing energy transition, essential for achieving carbon peaking and carbon neutrality goals. The proliferation of electric vehicles (EVs) introduces significant challenges to power distribution network stability due to their aggregated charging load in residential areas, particularly during peak electricity consumption periods. This paper proposes a method to predict the spatiotemporal distribution of EV charging demand in residential areas using geographic information points of interest (POI) data features and a decision‐making model. Utilizing real historical data, probability distribution models for EV users' arrival times and charging characteristics were constructed using Gaussian Mixture Models (GMM). The spatiotemporal characteristics of EV travel and charging behaviors were analyzed, and a comprehensive charging decision model incorporating both emergency and stochastic scenarios was developed. The model's efficacy in capturing the probability distributions of characteristic variables was validated through a case study. The results demonstrate the model's potential for accurately predicting EV charging demand, providing valuable insights for infrastructure planning and resource allocation. © 2024 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

Identifying flying insects is a significant challenge for biologists. Entomological lidar offers a unique solution, enabling rapid identification and classification in field settings. No other method can match its speed and efficiency in identifying insects in flight. This non-intrusive tool is invaluable for assessing insect biodiversity, informing conservation planning, and evaluating efforts to address declining insect populations. Although the species richness of co-existing insects can reach tens of thousands, current photonic sensors and lidars can differentiate roughly one hundred signal types. While the retrieved number of clusters correlate with Malaise trap diversity estimates, this taxonomic specificity, the number of discernible signal types is currently limited by instrumentation and algorithm sophistication. In this study, we report 32,533 observations of wild flying insects along a 500-meter transect. We report the benefits of lidar polarization bands for differentiating species and compare the performance of two unsupervised clustering algorithms, namely Hierarchical Cluster Analysis and Gaussian Mixture Model. Our analysis shows that polarimetric properties could be partially predicted even with unpolarized light, thus polarimetric lidar bands provide only a minor improvement in specificity. Finally, we use the physical properties of the clustered observations, such as wing beat frequency, daily activity patterns, and spatial distribution, to establish a lower bound for the number of species represented by the differentiated signal types. [ABSTRACT FROM AUTHOR]

The editorial discusses recent advances in deep learning for traffic probabilistic prediction, emphasizing the importance of probabilistic models in handling uncertainties in traffic systems. Deep learning methods such as Bayesian Deep Learning, Deep Gaussian Processes, Quantile Regression, and diffusion models are explored for their effectiveness in providing a range of possible traffic states and their corresponding probabilities. The practical implications of probabilistic traffic prediction include proactive congestion control, refined passenger information systems, improved risk assessment for autonomous driving, and enhanced public transport planning. Future research directions focus on developing efficient and scalable deep Bayesian inference techniques, integrating various data sources for more precise predictions, and incorporating uncertainty estimates into decision-making frameworks for robust traffic management solutions. [Extracted from the article]

The intelligibility of speech is a primary component to assess the severity level of Dysarthria, a speech disorder, which is caused not only due to weakness in vocal motor muscles but also difficulty in controlling its movements. Prior information about the severity of Dysarthria, aids the therapist during the rehabilitation process. This paper introduces a novel hybrid architecture comprising Gaussian Mixture Model and Neural Network (GMM-NN) for categorizing Dysarthric severity into four classes based on speech intelligibility. Mel Frequency Cepstral Coefficients (MFCC) extracted from the segmented speech signals are used to train the classifier. The proposed model produced a 1.9% improvement in accuracy when compared to the baseline Gaussian Mixture Model (GMM). The Gaussian Mixture Model Deep Neural Network (GMM-DNN) and Gaussian Mixture Model Feed Forward Neural Network (GMM-FFNN) architectures showed an accuracy of 96.7% and 96.42% with F1 scores of 0.9649, 0.9604 respectively. [ABSTRACT FROM AUTHOR]

A novel methodology to investigate the fine structure of interest rates based on Machine Learning techniques is discussed. The aim is to capture in an unsupervised way the common stochastic structure that drives the dynamics of interest rates of different maturities. The proposed approach is based on the Wasserstein barycenter, a powerful tool of analysis that allows us to construct, from a set of assigned probability distributions, a single probability distribution that captures the essential features of the whole set. To identify common stochastic factors, a Gaussian Mixture Model is fitted to the Wasserstein barycenter by maximum likelihood using the Expectation-Maximization algorithm with an initialization strategy based on Graph Machine Learning techniques. A fine-tuning of single-maturity interest rates is discussed in an attempt to capture maturity-specific stochastic factors. The proposed analysis also gives us the opportunity to test the hypothesis of a market segmentation into a short-term segment, the money market, and a long-term segment, the capital market, each with its own segment-specific stochastic factors. The methodology is tested on the US zero-coupon Treasury yield curve. The results obtained seem to show that most of the stochastic nature of the dynamics of the US zero-coupon yield curve can be captured by a three-component Gaussian Mixture Model describing the Wasserstein barycenter of the short-term segment of the yield curve. [ABSTRACT FROM AUTHOR]

The occurrence of accidents in expressway weaving areas is significantly influenced by frequent lane change maneuvers. Acquiring the lane change behavior pattern characteristics of vehicles in this area can provide prior knowledge for autonomous vehicles when performing lane change maneuvers, which helps ensure the safety of autonomous vehicles. This study aims to extract lane change behavior patterns of vehicles in weaving areas, to analyze the distribution differences of patterns across different lane change maneuvers, and to explore risk characteristics during the lane change process. First, a lane-changing sequence segmentation method was designed based on the hierarchical Dirichlet process hidden semi-Markov model (HDP-HSMM) algorithm, taking into account the interaction with surrounding vehicles and risk factors. Second, the Gaussian mixture model latent Dirichlet allocation (GMM-LDA) algorithm was employed to cluster the segments and derive patterns of lane-changing behavior that include risk attributes. The trajectory data from the UCF SST dataset were used to validate the method framework and make an in-depth analysis. The results show that the behavior patterns obtained by this method are able to better describe the operational and risk states of the vehicle. Variations exist in the behavioral patterns of different types of lane change maneuvers throughout the entire process. Spatial distribution disparities exist in the behavior patterns of lane change maneuvers across various sections of weaving areas. The findings of this study provide behavioral characteristics of different types of lane change maneuvers in weaving areas, which might contribute to enhancing the accurate recognition of lane change behaviors by autonomous vehicles. [ABSTRACT FROM AUTHOR]

Disaster resilience is an emerging concept for managing the risk of civil structural systems considering not only structural safety against extreme loads but also aftermath recovery efforts. The system reliability–based resilience analysis framework facilitates the quantitative evaluation of the disaster resilience performance of structural systems by assessing reliability and redundancy for numerous disruption scenarios. However, its practical application is limited due to the substantial number of structural analyses required to estimate the reliability and redundancy indices. To address the computational challenges, this paper proposes a new importance sampling algorithm, termed Importance Sampling for Noteworthy Scenarios (ISNS). The ISNS algorithm leverages a Gaussian process–based principal point search method to identify initial disruption scenarios that dominantly impact the structure's resilience. Subsequently, a mixture-based distribution is constructed to represent the near-optimal importance sampling density characterizing the failure domains of the noteworthy disruption scenarios. The two-step procedure enables the simultaneous estimation of both reliability and redundancy indices of the identified noteworthy scenarios. Furthermore, an active learning scheme is incorporated to efficiently train surrogates. Numerical examples of engineering applications are investigated to demonstrate the improved efficiency offered by the proposed method. However, the proposed method faces limitations in multihazard contexts and high-dimensional, highly nonlinear scenarios. These limitations necessitate further validation of the Gaussian process and mixture models to ensure robustness. [ABSTRACT FROM AUTHOR]

Limitations in instrument resolution mean that soil samples analysis using energy-dispersive X-ray fluorescence (EDXRF) often produced overlapping spectral peaks of heavy metals with similar energies. This overlapping of spectral peaks posed a significant problem in the quantitative analysis of heavy metals in soil samples. Therefore, it was necessary to decompose the overlapping peaks. In this paper, a new method for decomposing overlapping peaks was presented in the context of Gaussian Mixture Model (GMM) parameter estimation, combining the k-means algorithm and the African Vulture Optimization Algorithm (AVOA). The data acquired from EDXRF instruments were preprocessed before being input to the k-means algorithm, which produces the output used to initialize the parameters of the GMM. Subsequently, AVOA was employed to optimize the GMM parameters based on a fitness function derived from the data. Ultimately, the optimized GMM parameters obtained through the AVOA process yield the optimal solution for the decomposition of overlapping peaks. This approach effectively addressed the challenge of peak overlap in the determination of heavy metals in soil by EDXRF. Soil samples were obtained from national standard samples, peach gardens, and vegetable-growing regions. The results showed that the maximum absolute errors for the peak channel and peak area were 1.46% and 6.37%, respectively. A comparison of the proposed model with others models indicated that the k-means-GMM-AVOA model exhibits the best performance in terms of speed, accuracy, and stability. [ABSTRACT FROM AUTHOR]

The widespread adoption of nonlinear power electronic devices in residential settings has significantly increased the stochasticity and uncertainty of power systems. The original load power data, characterized by numerous irregular, random, and probabilistic components, adversely impacts the predictive performance of deep learning techniques, particularly neural networks. To address this challenge, this paper proposes a time-series probabilistic load power prediction technique based on the mature neural network point prediction technique, i.e., decomposing the load power data into deterministic and stochastic components. The deterministic component is predicted using deep learning neural network technology, the stochastic component is fitted with Gaussian mixture distribution model and the parameters are fitted using great expectation algorithm, after which the stochastic component prediction data is obtained using the stochastic component generation method. Using a mature neural network point prediction technique, the study evaluates six different deep learning methods to forecast residential load power. By comparing the prediction errors of these methods, the optimal model is identified, leading to a substantial improvement in prediction accuracy. [ABSTRACT FROM AUTHOR]

Copyright of Journal of South-Central Minzu University (Natural Science Edition) is the property of Journal of South-Central Minzu University (Natural Science Edition) Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Seismic fragility analysis and resilience assessment of large‐span cable‐stayed bridge structures are critical for evaluating their seismic performance. However, there is a scarcity of research on the effects of multi‐support ground motions and their non‐Gaussian characteristics on seismic fragility and resilience. This paper aims to addresses this issue. Initially, random ground motions with spatial variability and non‐Gaussian characteristics are simulated using the Spectral Representation Method (SRM) and the Unified Hermite Polynomial Model (UHPM). Subsequently, the Fractional Exponential Moments‐based Maximum Entropy Method (FEM‐MEM) and the Adaptive Gaussian Mixture Model (AGMM) are employed for seismic reliability‐based fragility analysis, overcoming the shortcomings of conventional lognormal assumption. Component‐ and system‐level fragility analyses are conducted sequentially, followed by seismic resilience assessment of bridge structures based on the results of system‐level fragility analysis. A numerical example is presented to validate the proposed method. Computational results indicate that: (1) The proposed method offers higher accuracy and broader applicability for seismic fragility analysis of large‐span cable‐stayed bridge structures compared to traditional assumptions. (2) The non‐Gaussian characteristics of ground motions may significantly impact the seismic fragility analysis and resilience assessment of large‐span bridge structures. [ABSTRACT FROM AUTHOR]

AbstractCausal mediation, pleiotropy, and replication analyses are three highly popular genetic study designs. Although these analyses address different scientific questions, the underlying statistical inference problems all involve large-scale testing of composite null hypotheses. The goal is to determine whether all null hypotheses - as opposed to at least one - in a set of individual tests should simultaneously be rejected. Recently, various methods have been proposed for each of these situations, including an appealing two-group empirical Bayes approach that calculates local false discovery rates (lfdr). However, lfdr estimation is difficult due to the need for multivariate density estimation. Furthermore, the multiple testing rules for the empirical Bayes lfdr approach can disagree with conventional frequentist z-statistics, which is troubling for a field that ubiquitously utilizes summary statistics. This work proposes a framework to unify two-group testing in genetic association composite null settings, the conditionally symmetric multidimensional Gaussian mixture model (csmGmm). The csmGmm is shown to demonstrate more robust operating characteristics than recently-proposed alternatives. Crucially, the csmGmm also offers interpretability guarantees by harmonizing lfdr and z-statistic testing rules. We extend the base csmGmm to cover each of the mediation, pleiotropy, and replication settings, and we prove that the lfdr z-statistic agreement holds in each situation. We apply the model to a collection of translational lung cancer genetic association studies that motivated this work. [ABSTRACT FROM AUTHOR]

The broader aim of this study is the corpus-based investigation of the written language production process. To this end, temporal markers have been keylog recorded alongside the writing processes to exploit pauses to segment the speech product into linear units of performance. However, identifying these pauses requires selecting the relevant interkey intervals (IKI) between production microevents. Different models have been applied to identify such components in the empirical distribution of these delays, in particular Gaussian mixture models (GMM), yet no consensus has emerged regarding the number and interpretation of these components. Here, we analyze IKI distributions from a corpus of keylogs and show that a model with two components is robustly selected across a large range of participants. Furthermore, we show that the contents of these modes are consistent with a fluency vs. disfluency interpretation of the modes by identifying which kinds of events fall into each mode. [ABSTRACT FROM AUTHOR]

Summary Obstructive sleep apnea is a heterogeneous sleep disorder with varying phenotypes. Several studies have already performed cluster analyses to discover various obstructive sleep apnea phenotypic clusters. However, the selection of the clustering method might affect the outputs. Consequently, it is unclear whether similar obstructive sleep apnea clusters can be reproduced using different clustering methods. In this study, we applied four well‐known clustering methods: Agglomerative Hierarchical Clustering; K‐means; Fuzzy C‐means; and Gaussian Mixture Model to a population of 865 suspected obstructive sleep apnea patients. By creating five clusters with each method, we examined the effect of clustering methods on forming obstructive sleep apnea clusters and the differences in their physiological characteristics. We utilized a visualization technique to indicate the cluster formations, Cohen's kappa statistics to find the similarity and agreement between clustering methods, and performance evaluation to compare the clustering performance. As a result, two out of five clusters were distinctly different with all four methods, while three other clusters exhibited overlapping features across all methods. In terms of agreement, Fuzzy C‐means and K‐means had the strongest (κ = 0.87), and Agglomerative hierarchical clustering and Gaussian Mixture Model had the weakest agreement (κ = 0.51) between each other. The K‐means showed the best clustering performance, followed by the Fuzzy C‐means in most evaluation criteria. Moreover, Fuzzy C‐means showed the greatest potential in handling overlapping clusters compared with other methods. In conclusion, we revealed a direct impact of clustering method selection on the formation and physiological characteristics of obstructive sleep apnea clusters. In addition, we highlighted the capability of soft clustering methods, particularly Fuzzy C‐means, in the application of obstructive sleep apnea phenotyping. [ABSTRACT FROM AUTHOR]

Background: Using neuroimaging techniques, growing evidence has suggested that the choroid plexus (CP) volume is enlarged in multiple neurodegenerative diseases, including amyotrophic lateral sclerosis (ALS). Notably, the CP has been suggested to play an important role in inflammation-induced CNS damage under disease conditions. However, to our knowledge, no study has investigated the relationships between peripheral inflammation and CP volume in sporadic ALS patients. Thus, in this study, we aimed to verify CP enlargement and explore its association with peripheral inflammation in vivo in independent ALS cohorts. Methods: Based on structural MRI data, CP volume was measured using Gaussian mixture models and further manually corrected in two independent cohorts of sporadic ALS patients and healthy controls (HCs). Serum inflammatory protein levels were measured using a novel high-sensitivity Olink proximity extension assay (PEA) technique. Xtreme gradient boosting (XGBoost) was used to explore the contribution of peripheral inflammatory factors to CP enlargement. Then, partial correlation analyses were performed. Results: CP volumes were significantly higher in ALS patients than in HCs in the independent cohorts. Compared with HCs, serum levels of CRP, IL-6, CXCL10, and 35 other inflammatory factors were significantly increased in ALS patients. Using the XGBoost approach, we established a model-based importance of features, and the top three predictors of CP volume in ALS patients were CRP, IL-6, and CXCL10 (with gains of 0.24, 0.18, and 0.15, respectively). Correlation analyses revealed that CRP, IL-6, and CXCL10 were significantly associated with CP volume in ALS patients (r = 0.462 ∼ 0.636, p < 0.001). Conclusion: Our study is the first to reveal a consistent and replicable contribution of peripheral inflammation to CP enlargement in vivo in sporadic ALS patients. Given that CP enlargement has been recently detected in other brain diseases, these findings should consider extending to other disease conditions with a peripheral inflammatory component. [ABSTRACT FROM AUTHOR]

Numerical simulations indicate that correlations exist between the velocity distributions of stars and dark matter (DM). We study the local DM velocity distribution based on these correlations. We select K giants from LAMOST DR8 cross-matched with Gaia DR3, which have robust measurements of velocity and metallicity, and separate them into the disk, halo substructure, and isotropic halo components in the chemodynamical space utilizing the Gaussian mixture model. The substructure component is highly radially anisotropic and possibly related to the Gaia–Enceladus–Sausage (GES) merger event, while the isotropic halo component is accreted from the earliest mergers following the Maxwell–Boltzmann distribution (standard halo model [SHM]). We find that the GES-like substructure contributes ∼85% of the local nondisk stars in the solar neighborhood, which is nearly invariant when applying different volume cuts or additional angular momentum constraints. Utilizing the metallicity–stellar mass relation and the stellar mass–halo mass relation, we find that ∼ 25 − 15 + 24 % of local DM is in the kinematic substructure. Combined with the stellar distributions of nondisk components, we modify the heliocentric velocity distribution of local DM. It shifts to a lower speed with a sharper peak compared to the SHM and updates the detection limits of DM direct detection experiments. We discuss extensively the degeneracies present in the GMM fitting and propose that more kinematic and chemical information such as α abundance could help to break the degeneracy in the future. Our work confirms that the local DM velocity distribution deviates significantly from the SHM and needs to be properly accounted for in the DM detection experiments. [ABSTRACT FROM AUTHOR]

Paddy is a major nutritional requirement for the human population. However, diseases such as bacterial leaf blight, brown spot, and leaf smut are significantly impacting the leaves of the paddy at various stages, resulting in significant losses. Consequently, this leads to an important decrease in production. An effective approach would involve assigning agricultural specialists to the affected area to provide remedies for the affected crops. This task is both time-consuming and tedious. The latest technological advancements will provide alternative solutions in a more efficient and quicker manner. We employed a deep transfer learning algorithm to assess the accuracy of six frequently utilised transfer learning algorithms. In order to improve the accuracy of the model, we boosted the quality of the images using adaptive filtering and Gaussian Mixture Model (GMM) clustering technique. This allows for the simple segmentation of images into diseased, normal, and surrounding regions, resulting in more accurate classification. We applied a deep transfer learning method to evaluate the accuracy of six widely used transfer learning algorithms, specifically VGG, ResNet, Inception, MobileNet, DenseNet, and EfficientNet. The DenseNet model achieved an average accuracy of 94%, with a precision of 92%, recall of 95%, and an F1-score of 93%. To improve performance, we implemented an ensemble technique, which involves combining two or more models. The Ensemble Convolutional Neural Network (ECNN) was designed by using three pre-trained architectures: ResNet, DenseNet, and EfficientNet.The ensemble model's final predict was determined by employing a weighted average methodology to enhance the prediction's accuracy. This technique shows a perfect balance between performance and accuracy in diagnosing diseases in paddy leaves. [ABSTRACT FROM AUTHOR]

Due to different acquisition time, viewpoint, and sensor noise during the process of point cloud data acquisition, the captured point clouds typically exhibit partial overlapped and contain large amounts of noise and outliers. However, this circumstance tends to diminish the accuracy of point-to-point correspondence searches. Existing point-level methods rely on idealized point-to-point correspondences, which cannot be guaranteed in practical applications. To address above limitations, a noval network based on a geometry transformer and a Gaussian Mixture Model (GMM) is proposed, called GTGMM. Specifically, we formulate the registration problem as the problem of aligning the two Gaussian mixtures, leveraging the advantages of the statistic model and learned robust features to overcome the noise and outliers variants. We utilize a Local Feature Extractor (LFE) to extract structural features of point clouds, while the Transformer encoders establish global relations among the point clouds. Additionally, a geometry transformer network is introduced to capture geometric relations within the point cloud, and overlap scores are learned to reject non-overlapping regions. Utilizing overlap scores, point cloud features, and 3D point cloud coordinates, the matching parameters of GMM to calculate to guide the alignment of two point clouds. Experimental results on synthetic datasets and the real Terracotta Warriors data demonstrate that our method achieves high accuracy and robustness under various registration conditions. [ABSTRACT FROM AUTHOR]

Physical unclonable functions (PUFs) have emerged as a favorable hardware security primitive, they exploit the process variations to provide unique signatures or secret keys amidst other critical cryptographic applications. CMOS-based PUFs are the most popular type, they generate unique bit strings using process variations in semiconductor fabrication. However, most existing CMOS PUFs are found to be vulnerable to modeling attacks based on machine learning (ML) algorithms. Memristors leveraging nanotechnology fabrication processes and highly nonlinear behavior became an interesting alternative to the existing CMOS-based PUF technology, introducing cryptographic and resilient random outputs. Memristor-based PUFs are emerging due to the inherent randomness at both the memristor level due to the cycle-to-cycle (C2C) programming variation of the device and the fabrication process level such as the cross-sectional area and variations. Our study focuses on building a machine learning analysis and attack framework of tools on C u / H f O 2 - x / p + + S i memristor-based PUF (MR-PUF). Our objective is to test the resiliency of the security margins of the presented PUF using machine learning analysis tools, on-top of holistic NIST cryptographic randomness testing initially provided, to provide a high level of certainty in predicting the randomness output of the verified Memrister-based PUF. Our main contribution is a holistic study that focuses on attacking the randomness output resiliency based on building randomness predictors using Logistic Regression (LR), Support Vector Machine (SVM), Gaussian Mixture Models (GMM), K-means, K-means + + , Random Forest, XGBoost and LSTM, within efficient time, and data complexity. Our results yield low accuracy and ROC results of within 0.49 - 0.52 and 0.49 - 0.52 respectively, indicating failure in predicting random data demonstrates efficient randomness prediction resiliency of the MR-PUF. The efficient time and data complexities of these attacks illustrated in this study are yielded to be linear and quadratic resulting in attack execution time in seconds and 5032 training samples combined with 2157 testing samples to verify the randomness of PUF. [ABSTRACT FROM AUTHOR]

To enhance the performance of non-orthogonal multiple access (NOMA)-assisted integrated sensing and communication (ISAC) systems in multi-user distributed scenarios, an improved Gaussian Mixture Model (GMM)-based user clustering algorithm is proposed. This algorithm is tailored for ISAC systems, significantly improving bandwidth reuse gains and reducing serial interference. First, using the Sum of Squared Errors (SSE), the algorithm reduces sensitivity to the initial cluster center locations, improving clustering accuracy. Then, direction weight factors are introduced based on the base station position and a penalty function involving users' Euclidean distances and sensing power. Modifications to the EM algorithm in calculating posterior probabilities and updating the covariance matrix help align user clusters with the characteristics of NOMAISAC systems. This improves users' interference resistance, lowers decoding difficulty, and optimizes the system's sensing capabilities. Finally, a fractional programming (FP) approach addresses the non-convex joint beamforming design problem, enhancing power and channel gains and achieving co-optimizing sensing and communication signals. The simulation results show that, under the improved GMM user clustering algorithm and FP optimization, the NOMA-ISAC system improves user spectral efficiency by 4.3% and base station beam intensity by 5.4% compared to traditional ISAC systems. [ABSTRACT FROM AUTHOR]

AbstractClustering is a widely used technique with a long and rich history in a variety of areas. However, most existing algorithms do not scale well to large datasets, or are missing theoretical guarantees of convergence. This paper introduces a provably robust clustering algorithm based on loss minimization that performs well on Gaussian mixture models with outliers. It provides theoretical guarantees that the algorithm obtains high accuracy with high probability under certain assumptions. Moreover, it can also be used as an initialization strategy for k-means clustering. Experiments on real-world large-scale datasets demonstrate the effectiveness of the algorithm when clustering a large number of clusters, and a k-means algorithm initialized by the algorithm outperforms many of the classic clustering methods in both speed and accuracy, while scaling well to large datasets such as ImageNet. [ABSTRACT FROM AUTHOR]

With the vigorous development of building structures and important infrastructure, structural health monitoring is necessary. Because there is no need to establish structural finite element modeling and train for various structural conditions, the data-driven and unsupervised learning method is very popular. Principal component analysis is a powerful signal analysis tool, but its lack of physical significance and the loss of sensitive information have hindered its wider application. Therefore, the improved principal component analysis based narrowband filtering is proposed to extract mode shapes and construct the structural state vectors, so that the damage index is more sensitive to damage and robust to the environmental factors. After the vibration response of the long-term monitoring is analyzed by the principal component analysis, the Gaussian mixture model clustering analysis is used to classify the structural states. Finally, the proposed method is applied to the analysis of the simulation data of ASCE Benchmark structure and the measured data of steel beams in the lab. The results show that the structural state vector is sensitive to structural damage. The clustering analysis of Gaussian mixture model can distinguish the structural states. The effectiveness of the proposed method is verified. [ABSTRACT FROM AUTHOR]

Common Variable Immunodeficiency (CVID) is a primary immunodeficiency characterized by reduced levels of specific immunoglobulins, resulting in frequent infections, autoimmune disorders, increased cancer risk, and diminished antibody production despite an adequate B cell count. With its clinical manifestations being highly variable, the classification of CVID, including the widely recognized Freiburg classification, is primarily based on clinical symptoms and genetic variations. Our study aims to refine the classification of CVID by analyzing transcriptomics data to identify distinct disease subtypes. We utilized the GSE51405 dataset, examining transcriptomic profiles from 30 CVID patients without complications. Employing a combination of clustering techniques—KMeans, hierarchical agglomerative clustering, spectral clustering, and Gaussian Mixture models—and differential gene expression analysis with R's limma package, we integrated molecular findings with demographic data (age and gender) through correlation analysis and identified common genes among clusters. Three distinct clusters of CVID patients were identified using KMeans, Agglomerative Clustering, and Gaussian Mixture Models, highlighting the disease's heterogeneity. Differential expression analysis unveiled 31 genes with variable expression levels across these clusters. Notably, nine genes (EIF5A, RPL21, ANP32A, DTX3L, NCF2, CDC42EP3, CHP1, FOLR3, and DEFA4) exhibited consistent differential expression across all clusters, independent of demographic factors. The study recommends categorizing patients based on the four genes, NCF2, CHP1, FOLR3, and DEFA4—as they may assist in prognostic prediction. Transcriptomic analysis of common variable immunodeficiency (CVID) patients identified three distinct clusters based on gene expression, independent of age and gender. Nine differentially expressed genes were identified across these clusters, suggesting potential biomarkers for CVID subtype classification. These findings highlight the genetic heterogeneity of CVID and provide novel insights into disease classification and potential personalized treatment approaches. [ABSTRACT FROM AUTHOR]

Fluvial sediment dynamics in mountain rivers are changing rapidly in a degrading cryosphere, raising the potential for erosive rainfall and runoff and detrimental effects on downstream areas. Hence, we need to understand better what characterises and drives episodic pulses of water and suspended solids in rivers. Here, we infer different types of such sediment-discharge events from 959 automatically detected events based on 16 metrics derived from 15 min time series of streamflow and suspended sediment concentrations from Vent–Rofental in the high Ötztal Alps, Austria. We use principal component analysis to extract uncorrelated event characteristics and cluster event types with a Gaussian mixture model. We interpret the thus inferred event types with catchment metrics describing antecedent conditions, hydrometeorological forcing, and fraction of catchment area with freezing temperatures and snow cover. We find event magnitude, hysteresis, and event shape complexity to be the main factors characterising the overall event regime. The most important characteristics distinguishing the event types are suspended sediment and streamflow magnitude and complexity of the hydro- and sedigraphs. Sediment-discharge hysteresis is less relevant for discerning event types. We derive four event types that we attribute to (1) compound rainfall–melt extremes, (2) glacier and seasonal snowmelt, (3) freeze–thaw-modulated snowmelt and precipitation events, and (4) late-season glacier melt. Glacier and snowmelt events driven by warm conditions and high insolation were the most frequent and contributed some 40 % to annual suspended sediment yield on average; compound rainfall–melt extremes were the rarest but contributed the second-highest proportion (26 %). Our approach represents a reproducible method for objectively estimating the variety of event-scale suspended sediment transport conditions in mountain rivers, which can provide insights into the contribution of different drivers to annual sediment yields in current and future regimes. Our findings highlight the importance of both meltwater and rainfall–runoff as drivers of high-magnitude suspended sediment fluxes in mountain rivers. [ABSTRACT FROM AUTHOR]

We study the instantaneous inference of an unbounded planar flow from sparse noisy pressure measurements. The true flow field comprises one or more regularized point vortices of various strength and size. We interpret the true flow's measurements with a vortex estimator, also consisting of regularized vortices, and attempt to infer the positions and strengths of this estimator assuming little prior knowledge. The problem often has several possible solutions, many due to a variety of symmetries. To deal with this ill posedness and to quantify the uncertainty, we develop the vortex estimator in a Bayesian setting. We use Markov-chain Monte Carlo and a Gaussian mixture model to sample and categorize the probable vortex states in the posterior distribution, tailoring the prior to avoid spurious solutions. Through experiments with one or more true vortices, we reveal many aspects of the vortex inference problem. With fewer sensors than states, the estimator infers a manifold of equally possible states. Using one more sensor than states ensures that no cases of rank deficiency arise. Uncertainty grows rapidly with distance when a vortex lies outside of the vicinity of the sensors. Vortex size cannot be reliably inferred, but the position and strength of a larger vortex can be estimated with a much smaller one. In estimates of multiple vortices their individual signs are discernible because of the nonlinear coupling in the pressure. When the true vortex state is inferred from an estimator of fewer vortices, the estimate approximately aggregates the true vortices where possible. [ABSTRACT FROM AUTHOR]

Traffic accidents occur frequently in cold and snow‐ or ice‐covered regions due to weather changes that occur during the winter season. To detect the snow‐ or ice‐covered roads in road surface conditions, road surface images captured using fixed‐point cameras installed along the route are sufficient. This paper proposes a snow‐ or ice‐covered road detection method that uses the deep convolutional autoencoding Gaussian mixture model (DCAGMM) with structural similarity (SSIM). The DCAGMM method, which is an unsupervised anomaly detection method, is unaffected by imbalance in the training data. In addition, the end‐to‐end convolutional neural network implemented in the DCAGMM enables the capture of the unique characteristics of the road surface images. Finally, by reconstructing the input images as normal images, the comparison of the input and reconstructed images enables identification of snow‐ or ice‐covered road areas without requiring pixel‐level annotations. Furthermore, the road surface images include complex characteristics for reconstruction, and the SSIM‐based reconstruction error allows us to preserve the image quality of the reconstructed image. Experimental results obtained on real‐world road surface images demonstrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Water classification in Synthetic Aperture Radar (SAR) images is an ongoing area of research, which has implications in environmental monitoring and water resource management. Adaptive threshold algorithms provide a fast, reliable and efficient way to perform automated water classification, but users often lack awareness on selecting the best algorithm for their specific application. This paper presents a comprehensive assessment of adaptive threshold algorithms for water delineation applied to L- and C-band SAR backscatter images. We introduce a novel approach for dynamic selection of windows within a SAR image to determine optimum thresholds on sigma naught values. A comparison of five threshold-determination techniques is performed which include Otsu, Kittler and Illingworth (KI), Gaussian Mixture Model (GMM), Quality Index (QI) and Gamma Maximum Likelihood Estimation (GMLE) algorithms. We observed that, for L-band SAR data, convex hull approach produced better kappa coefficient value with GMM, KI and GMLE algorithms. However, for C-band SAR, kappa coefficients were highest for convex hull method with GMM, KI, QI and GMLE approaches and noticeably higher (> 0.89) when compared to split window approach. Our analysis indicates that the proposed convex hull method for window selection performs better in both L- and C-band SAR images. The results of our analysis will help users in identifying the best adaptive algorithm for water delineation in L- and C-band SAR images. [ABSTRACT FROM AUTHOR]

Copyright of Coal Geology & Exploration is the property of Xian Research Institute of China Coal Research Institute and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Beijing University of Technology is the property of Journal of Beijing University of Technology, Editorial Department and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Improving the autonomy of underwater interventions by remotely operated vehicles (ROVs) can help mitigate the impact of communication delays on operational efficiency. Currently, underwater interventions for ROVs usually rely on real-time teleoperation or preprogramming by operators, which is not only time-consuming and increases the cognitive burden on operators but also requires extensive specialized programming. Instead, this paper uses the intuitive learning from demonstrations (LfD) approach that uses operator demonstrations as inputs and models the trajectory characteristics of the task through the dynamic movement primitive (DMP) approach for task reproduction as well as the generalization of knowledge to new environments. Unlike existing applications of DMP-based robot trajectory learning methods, we propose the underwater DMP (UDMP) method to address the problem that the complexity and stochasticity of underwater operational environments (e.g., current perturbations and floating operations) diminish the representativeness of the demonstrated trajectories. First, the Gaussian mixture model (GMM) and Gaussian mixture regression (GMR) are used for feature extraction of multiple demonstration trajectories to obtain typical trajectories as inputs to the DMP method. The UDMP method is more suitable for the LfD of underwater interventions than the method that directly learns the nonlinear terms of the DMP. In addition, we improve the commonly used homomorphic-based teleoperation mode to heteromorphic mode, which allows the operator to focus more on the end-operation task. Finally, the effectiveness of the developed method is verified by simulation experiments. [ABSTRACT FROM AUTHOR]

Machine learning (ML) has been used in human gait data for appropriate assistive device prediction. However, their uptake in the medical setup still remains low due to their black box nature which restricts clinicians from understanding how they operate. This has led to the exploration of explainable ML. Studies have recommended local interpretable model‐agnostic explanation (LIME) because it builds sparse linear models around an individual prediction in its local vicinity hence fast and also because it could be used on any ML model. LIME is however, is not always stable. The research aimed to enhance LIME to attain stability by avoid the sampling step through combining Gaussian mixture model (GMM) sampling. To test performance of the GMM‐LIME, supervised ML were recommended because study revealed that their accuracy was above 90% when used on human gait. Neural networks were adopted for GaitRec dataset and Random Forest (RF) was adopted and applied on HugaDB datasets. Maximum accuracies attained were multilayer perceptron (95%) and RF (99%). Graphical results on stability and Jaccard similarity scores were presented for both original LIME and GMM‐LIME. Unlike original LIME, GMM‐LIME produced not only more accurate and reliable but also consistently stable explanations. [ABSTRACT FROM AUTHOR]

Data quality, especially label quality, may have a significant impact on the prediction accuracy in supervised learning. Training on datasets with label noise causes a degradation in performance and a reduction in prediction accuracy. To overcome the numerical label noise problem in regression, we estimate the posterior distribution of the true label through the Gaussian mixture model (GMM). Then, label noise estimation is proposed by integrating the idea of maximum a posteriori (MAP) estimation with the posterior distribution. Besides, a noise filtering algorithm with MAP estimation (MAPNF) is designed by combining the optimal sample selection framework with the estimator. Extensive experiments are carried out on benchmark datasets and an age estimation dataset to verify the effectiveness of MAPNF. The results on benchmark datasets show that MAPNF outperforms other latest filtering algorithms in improving the generalization performance of different regression models, including noise-sensitive models and noise-robust models. The model error can be reduced by 29.7% to 69.6%. Our proposed approach can also identify erroneous labels in an age estimation dataset (total of 18424). The model trained on the filtered dataset (19% of the data removed) achieves a reduced test error on the dataset by at least 2.68%. The results demonstrate a less-is-better effect by achieving lower prediction errors with fewer high-quality samples. It can be concluded that MAPNF can effectively identify label noise and optimize the data quality. [ABSTRACT FROM AUTHOR]

Future 6G communications are envisioned to enable a large catalogue of pioneering applications. These will range from networked Cyber-Physical Systems to edge computing devices, establishing real-time feedback control loops critical for managing Industry 5.0 deployments, digital agriculture systems, and essential infrastructures. The provision of extensive machine-type communications through 6G will render many of these innovative systems autonomous and unsupervised. While full automation will enhance industrial efficiency significantly, it concurrently introduces new cyber risks and vulnerabilities. In particular, unattended systems are highly susceptible to trust issues: malicious nodes and false information can be easily introduced into control loops. Additionally, Denial-of-Service attacks can be executed by inundating the network with valueless noise. Current anomaly detection schemes require the entire transformation of the control software to integrate new steps and can only mitigate anomalies that conform to predefined mathematical models. Solutions based on an exhaustive data collection to detect anomalies are precise but extremely slow. Standard models, with their limited understanding of mobile networks, can achieve precision rates no higher than 75%. Therefore, more general and transversal protection mechanisms are needed to detect malicious behaviors transparently. This paper introduces a probabilistic trust model and control algorithm designed to address this gap. The model determines the probability of any node to be trustworthy. Communication channels are pruned for those nodes whose probability is below a given threshold. The trust control algorithm comprises three primary phases, which feed the model with three different probabilities, which are weighted and combined. Initially, anomalous nodes are identified using Gaussian mixture models and clustering technologies. Next, traffic patterns are studied using digital Bessel functions and the functional scalar product. Finally, the information coherence and content are analyzed. The noise content and abnormal information sequences are detected using a Volterra filter and a bank of Finite Impulse Response filters. An experimental validation based on simulation tools and environments was carried out. Results show the proposed solution can successfully detect up to 92% of malicious data injection attacks. [ABSTRACT FROM AUTHOR]

In deep mineral exploration, it is difficult to constrain the complex geological structures using a single geophysical method. To tackle the difficulty, integrated geophysical surveys and joint data interpretation are essential. Machine learning (ML) provides more accurate predictions than traditional methods, especially when dealing with complex data from multiple sources or varied statistical distributions. However, a major challenge in using ML for deep mineral exploration is the scarcity and imbalance of labeled samples, mainly due to budget constraints and the complexity of ore deposits. This issue reduces the accuracy of predictive models and introduces bias. Additionally, limited labeling can lead to difficulties in predicting previously undefined classes in training datasets. To address these challenges, we introduce a robust semisupervised ML framework that integrates diverse geophysical and geological datasets to improve model reliability with limited labeled data. Our approach uses a semisupervised ML variational Gaussian mixture model (SsL-VGMM) to handle issues related to insufficient and imbalanced data. We enhanced the model's predictive capability for unseen data by introducing a novel penalty factor in the 'cannot-link' function. Moreover, we employed Bayesian optimization, focusing on the mean-mixture weight, to avoid local optima during model training. Our model demonstrated high accuracy and efficiency, with classification and prediction accuracies of 95.33% and 87.4%, respectively, in numerical and electromagnetic simulation scenarios. Its effectiveness was further validated by locating Pb–Zn–Ag deposits in Inner Mongolia, supported by actual drilling data. This paper highlights the model's potential in complex mineral exploration and its significant practical and innovative value for deep mineral exploration. [ABSTRACT FROM AUTHOR]

Deep Neural Networks (DNNs) have revolutionized various fields through their ability to model complex patterns, yet their performance critically hinges on the availability of large-scale, accurately labeled datasets. The costly and time-consuming process of creating such datasets has motivated research into Learning with Noisy Labels (LNL), which aims to reduce reliance on perfect labels. This paper introduces CoFix, an innovative LNL framework that integrates sample selection with semi-supervised learning techniques to address the challenge of noisy labels in training. CoFix employs a Gaussian Mixture Model (GMM) to dynamically segment the training data into clean and noisy subsets, leveraging semi-supervised learning approaches on both. Inspired by the FixMatch algorithm, CoFix refines consistent regularization and pseudo-labeling strategies, enhancing augmentation strategies and temperature sharpening techniques. Additionally, CoFix explores label smoothing to augment the loss function, further refining model performance. Our experiments demonstrate CoFix's superiority over state-of-the-art methods, achieving significant improvements in fewer training epochs, particularly in lower noise scenarios. The robustness and versatility of CoFix are evident through its consistent performance across various benchmark datasets and noise levels. The contributions of this paper include a novel LNL method with enhanced generalization capability, an investigation into the impact of label smoothing on the loss function, and extensive testing that confirms CoFix's efficiency and adaptability to different noise levels. [ABSTRACT FROM AUTHOR]

Context. The nuclear stellar disc (NSD) is a flat and dense stellar structure at the centre of the Milky Way. Previous work has identified the presence of metal-rich and metal-poor stars in the NSD, suggesting that they have different origins. The recent publication of photometric, metallicity, proper motion, and orbital catalogues allows the NSD stellar population to be characterised with unprecedented detail. Aims. We aim to explore the proper motions and orbits of NSD stars with different metallicities to assess whether they have different origins and to better understand the metallicity distribution in the NSD. Methods. We distinguished between metal-rich and metal-poor stars by applying a Gaussian mixture model, as done in previous work, and analysed the proper motions, orbits, and spatial distribution of stars with different metallicities. Results. We find that metal-rich stars exhibit a lower velocity dispersion, suggesting that they trace a kinematically cooler component compared to metal-poor ones. Furthermore, z-tube orbits are predominant among metal-rich stars, while chaotic/box orbits are more common among metal-poor ones. We also find that metal-rich and metal-poor stars show a similar extinction and are present throughout the analysed regions. As a secondary result, we detected a metallicity gradient in the metal-rich population with higher metallicity towards the centre of the NSD and a tentative gradient for the metal-poor stars, which is consistent with previous studies that did not distinguish between the two populations. Conclusions. Our results suggest that metal-rich stars trace the NSD, whereas metal-poor ones are related to the Galactic bar and probably constitute Galactic bar interlopers and/or are NSD stars that originated from accreted clusters. The detected metallicity gradients aligns with the currently accepted inside-out formation of the NSD. [ABSTRACT FROM AUTHOR]

Recent developments have highlighted the critical role that computer-aided diagnosis (CAD) systems play in analyzing whole-slide digital histopathology images for detecting gastric cancer (GC). We present a novel framework for gastric histology classification and segmentation (GHCS) that offers modest yet meaningful improvements over existing CAD models for GC classification and segmentation. Our methodology achieves marginal improvements over conventional deep learning (DL) and machine learning (ML) models by adaptively focusing on pertinent characteristics of images. This contributes significantly to our study, highlighting that the proposed model, which performs well on normalized images, is robust in certain respects, particularly in handling variability and generalizing to different datasets. We anticipate that this robustness will lead to better results across various datasets. An expectation-maximizing Naïve Bayes classifier that uses an updated Gaussian Mixture Model is at the heart of the suggested GHCS framework. The effectiveness of our classifier is demonstrated by experimental validation on two publicly available datasets, which produced exceptional classification accuracies of 98.87% and 97.28% on validation sets and 98.47% and 97.31% on test sets. Our framework shows a slight but consistent improvement over previously existing techniques in gastric histopathology image classification tasks, as demonstrated by comparative analysis. This may be attributed to its ability to capture critical features of gastric histopathology images better. Furthermore, using an improved Fuzzy c-means method, our study produces good results in GC histopathology picture segmentation, outperforming state-of-the-art segmentation models with a Dice coefficient of 65.21% and a Jaccard index of 60.24%. The model's interpretability is complemented by Grad-CAM visualizations, which help understand the decision-making process and increase the model's trustworthiness for end-users, especially clinicians. [ABSTRACT FROM AUTHOR]

Semi-supervised object detection aims to enhance object detectors by utilizing a large number of unlabeled images, which has gained increasing attention in natural scenes. However, when these methods are directly applied to scenes with tiny objects, they face the challenge of selecting pseudo-labels with high localization quality due to the minuscule and blurred characteristics of these objects. To address this issue, we propose a novel method called semi-supervised tiny object detection (STOD). Firstly, to enhance the localization accuracy of pseudo-labels, we design a dense IoU-aware head that evaluates the quality of bounding box localization by incorporating additional predicted overlap values. Secondly, to mine more potential pseudo-labels, we propose a GMM-based multi-threshold pseudo-labels mining module that dynamically generates multiple thresholds using classification scores and overlap values to classify bounding boxes into strong positive and weak positive pseudo-labels. Lastly, we design the localization-aware weighting loss to incorporate the localization quality of both positive and negative samples in order to enhance the accuracy of pseudo-label localization. The experimental results show that STOD achieves comparable performance when compared to both fully and semi-supervised methods. Notably, on the VisDrone-Partial benchmark, STOD achieves outstanding results by outperforming our baseline model with improvements of 3.1 mAP, 5.7 A P 50 , and 3.0 A P 75 . [ABSTRACT FROM AUTHOR]

The growing problem of unsolicited text messages (smishing) and data irregularities necessitates stronger spam detection solutions. This paper explores the development of a sophisticated model designed to identify smishing messages by understanding the complex relationships among words, images, and context-specific factors, areas that remain underexplored in existing research. To address this, we merge a UCI spam dataset of regular text messages with real-world spam data, leveraging OCR technology for comprehensive analysis. The study employs a combination of traditional machine learning models, including K-means, Non-Negative Matrix Factorization, and Gaussian Mixture Models, along with feature extraction techniques such as TF-IDF and PCA. Additionally, deep learning models like RNN-Flatten, LSTM, and Bi-LSTM are utilized. The selection of these models is driven by their complementary strengths in capturing both the linear and non-linear relationships inherent in smishing messages. Machine learning models are chosen for their efficiency in handling structured text data, while deep learning models are selected for their superior ability to capture sequential dependencies and contextual nuances. The performance of these models is rigorously evaluated using metrics like accuracy, precision, recall, and F1 score, enabling a comparative analysis between the machine learning and deep learning approaches. Notably, the K-means feature extraction with vectorizer achieved 91.01% accuracy, and the KNN-Flatten model reached 94.13% accuracy, emerging as the top performer. The rationale behind highlighting these models is their potential to significantly improve smishing detection rates. For instance, the high accuracy of the KNN-Flatten model suggests its applicability in real-time spam detection systems, but its computational complexity might limit scalability in large-scale deployments. Similarly, while K-means with vectorizer excels in accuracy, it may struggle with the dynamic and evolving nature of smishing attacks, necessitating continual retraining. [ABSTRACT FROM AUTHOR]

Our interest was to evaluate changes in fixation duration as a function of time-on-task (TOT) during a random saccade task. We employed a large, publicly available dataset. The frequency histogram of fixation durations was multimodal and modelled as a Gaussian mixture. For this specific task, we found five fixation types. The "ideal" response would be a single accurate saccade after each target movement, with a typical saccade latency of 200-250 msec, followed by a long fixation (> 800 msec) until the next target jump. We found fixations like this, but they comprised only 10% of all fixations and were the first fixation after target movement only 23.4% of the time. More frequently (57.4% of the time), the first fixation after target movement was short (117.7 msec mean) and was commonly followed by a corrective saccade. Across the entire 100 sec of the task, median total fixation duration decreased. This decrease was approximated with a power law fit with R2 = 0.94. A detailed examination of the frequency of each of our five fixation types over time on task (TOT) revealed that the three shortest duration fixation types became more and more frequent with TOT whereas the two longest fixations became less and less frequent. In all cases, the changes over TOT followed power law relationships, with R2 values between 0.73 and 0.93. We concluded that, over the 100 second duration of our task, long fixations are common in the first 15 to 22 seconds but become less common after that. Short fixations are relatively uncommon in the first 15 to 22 seconds but become more and more common as the task progressed. Apparently. the ability to produce an ideal response, although somewhat likely in the first 22 seconds, rapidly declines. This might be related to a noted decline in saccade accuracy over time. [ABSTRACT FROM AUTHOR]

Summary Bayesian mixture modelling is widely used for density estimation and clustering. The Dirichlet process mixture model (DPMM) is the most popular Bayesian non‐parametric mixture modelling approach. In this manuscript, we study the choice of prior for the variance or precision matrix when Gaussian kernels are adopted. Typically, in the relevant literature, the assessment of mixture models is done by considering observations in a space of only a handful of dimensions. Instead, we are concerned with more realistic problems of higher dimensionality, in a space of up to 20 dimensions. We observe that the choice of prior is increasingly important as the dimensionality of the problem increases. After identifying certain undesirable properties of standard priors in problems of higher dimensionality, we review and implement possible alternative priors. The most promising priors are identified, as well as other factors that affect the convergence of MCMC samplers. Our results show that the choice of prior is critical for deriving reliable posterior inferences. This manuscript offers a thorough overview and comparative investigation into possible priors, with detailed guidelines for their implementation. Although our work focuses on the use of the DPMM in clustering, it is also applicable to density estimation. [ABSTRACT FROM AUTHOR]

Train fault detection often relies on comparing collected images with reference images, making accurate image type recognition crucial. Current systems use Automatic Equipment Identification (AEI) devices to recognize carriage numbers while capturing images, but damaged Radio Frequency (RF) tags or blurred characters can hinder this process. Carriage linear array images, with their high resolution, extreme aspect ratios, and local nonlinear distortions, present challenges for recognition algorithms. This paper proposes a method tailored for recognizing such images. We apply an object detection algorithm to locate key components, simplifying image recognition into a sparse point set alignment task. To handle local distortions, we introduce a weighted radial basis function (RBF) and maximize the similarity between Gaussian mixtures of point sets to determine RBF weights. Experiments show 100% recognition accuracy under nonlinear distortions up to 15%. The algorithm also performs robustly with detection errors and identifies categories from 79 image classes in 24 ms on an i7 CPU without GPU support. This method significantly reduces system costs and advances automatic exterior fault detection for trains. [ABSTRACT FROM AUTHOR]

Image compression is a widely used technique to reduce the spatial redundancy in images. Recently, learning based image compression has achieved significant progress by using the powerful representation ability from neural networks. However, the current learning based image compression methods suffer from the huge computational cost, which limits their capacity for practical applications. In this paper, we propose a unified framework called Efficient Deep Image Compression (EDIC) based on three new technologies, including a channel attention module, a Gaussian mixture model and a decoder-side enhancement module. Specifically, we design an auto-encoder style network for learning based image compression. To improve the coding efficiency, we exploit the channel relationship between latent representations by using the channel attention module. Besides, the Gaussian mixture model is introduced for the entropy model and improves the accuracy for bitrate estimation. Furthermore, we introduce the decoder-side enhancement module to further improve image compression performance. Our EDIC method can also be readily incorporated with the Deep Video Compression (DVC) framework (Lu et al. 2019) to further improve the video compression performance. Simultaneously, our EDIC method boosts the coding performance significantly while bringing slightly increased computational cost. More importantly, experimental results demonstrate that the proposed approach outperforms the current image compression methods and is up to more than 150 times faster in terms of decoding speed when compared with Minnen's method (Minnen et al. 2018). Moreover, we also evaluate the performance of the human-centric task (i.e., face recognition) by using different coding strategies. [ABSTRACT FROM AUTHOR]

AbstractUnderstanding the various roles and their corresponding evolving patterns is essential for evaluating the health of learning communities, such as in Asynchronous Online Discussion (AOD) forums. Existing research primarily focuses on identifying specific roles without considering the dynamics of role evolution, which are significant to ongoing community development. Guided by the Reader-to-Leader Framework (RtLF), this study employs multiple machine learning approaches to identify and predict the evolution of roles over time in a learning community. Our approach was applied to a K-12 Math online learning community, using data from 51,821 students to evaluate the community’s health over two years. First, a Gaussian Mixture Model was used to cluster students into five groups that represent their roles roughly in RtLF using the first nine months of learning activities. We then trained an XGBoost model to classify learner roles for subsequent periods based on meta-data. Third, we evaluate the health of the community by comparing the distribution of different roles over time and argue it is a healthy community. This study demonstrates how machine learning approaches can identify learner roles and support the health and development of learning communities. [ABSTRACT FROM AUTHOR]

A vast number of mass flow controllers (MFCs) are used in semiconductor industry. For the stable supply, an efficient production method of MFC is required. The gain tuning of the proportional-integral (PI) control to realize a setting flow rate is essential for efficient mass production. The gains are tuned to meet the specifications required for evaluation indices of response time and overshoot amount in a step response waveform. The tuning is complicated especially for the case of pressure-based MFCs. In this paper, we propose a simple method for the PI gain tuning using the Gaussian mixture model (GMM) and the direct inverse analysis applicable to the pressure-based MFCs' production. The relationship between the gains and evaluation indices for a standard unit of the MFC is modeled as the GMM. The direct inverse analysis calculates the difference between the standard and a test unit. Under the assumption that the difference can be compensated by a simple shift, gains likely to meet the specifications for the test unit are searched. We applied the method to seven test units. The result showed that the gains of all the test units were tuned within only a few iterations whose numbers were much less than the conventional manual tuning method, and there was no untunable unit. [ABSTRACT FROM AUTHOR]