计算机科学最新文献

{"title":"Society Officers & Administrative Committee","authors":"","doi":"10.1109/MAP.2025.3607581","DOIUrl":"https://doi.org/10.1109/MAP.2025.3607581","url":null,"abstract":"","PeriodicalId":13090,"journal":{"name":"IEEE Antennas and Propagation Magazine","volume":"67 5","pages":"C3-C3"},"PeriodicalIF":5.7,"publicationDate":"2025-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11212641","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145335278","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Consistent Image Inpainting with Pre-Perception and Cross-Perception Collaborative Processes.","authors":"Yongle Zhang,Yimin Liu,Hao Fan,Ruotong Hu,Jian Zhang,Qiang Wu","doi":"10.1109/tip.2025.3622071","DOIUrl":"https://doi.org/10.1109/tip.2025.3622071","url":null,"abstract":"It has been proven that introducing multiple guidance sources boosts image inpainting performance. However, existing methods primarily focus on local relationships and neglect the holistic interplay between guidance and texture information. Moreover, they lack an effective feedback mechanism to adaptively update the guidance process as corrupted texture information is progressively restored, potentially resulting in inconsistent inpainting. To tackle this issue, we propose a novel scheme aligned with pre-perception and cross-perception collaborative processes in human drawing. To mimic the pre-perception process, we introduce a pre-perceptual transformer block that captures long-range contextual dependencies and activates meaningful information to individually optimize image structures, semantic layouts, and textures, thereby effectively controlling their respective generation. To mimic the cross-perception collaborative process, we propose a cyclic cross-perceptual interaction to maintain consistency across the entire image regarding structure, layout, and texture while progressively refining their details. This interaction accounts for the global attention relationship between texture and other guidance sources (including image structure and semantic layout) to enhance image texture, alongside integrating a dedicated feedback mechanism to update guidance information. The proposed components are alternately deployed in three-branch decoders of the new scheme from rough to fine-grained levels to achieve these two iterative processes of human drawing. Experimental results prove the superiority of the proposed scheme over state-of-the-art methods across three datasets.","PeriodicalId":13217,"journal":{"name":"IEEE Transactions on Image Processing","volume":"26 1","pages":""},"PeriodicalIF":10.6,"publicationDate":"2025-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145338605","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Machine learning solutions with deep multilayer exogenous networks for distributed denial of service attacks model on networked resources in critical infrastructure","authors":"Rana Abdullah Zaeem ,&nbsp;Chuan-Yu Chang ,&nbsp;Maryam Pervaiz Khan ,&nbsp;Muhammad Shoaib ,&nbsp;Chi-Min Shu ,&nbsp;Muhammad Asif Zahoor Raja","doi":"10.1016/j.engappai.2025.112872","DOIUrl":"10.1016/j.engappai.2025.112872","url":null,"abstract":"<div><div>The increasing dependency on critical infrastructure and the vulnerability to cyber-attacks, particularly Distributed Denial of Service attacks, pose significant challenges and threats in this cold warfare era. This paper explores an epidemic model based distributed denial of service attacks system to analyze the impact of seclusion strategies on protecting critical infrastructure against cyber-attacks by leveraging machine learning knowledge with non-linear exogenous networks supported with Levenberg-Marquardt backpropagation. The proposed information security model presents the critical infrastructure nodes into susceptible, infected, quarantined and recovered differential compartments for the targeted population to portray the attack's dynamics and quarantine measures effectively. To analyze the rates for infection, efficiency in the quarantine and the recovery state, the synthetic data is acquired to carry out processes on various scenarios with Adams numerical solver and the said information is fed to intelligent supervised nonlinear autoregressive exogenous neural networks to decipher the attack patterns. The efficacy of the proposed stochastic computing paradigm is established on mean squared error-based convergence trends, error in time series illustrations, error-histogram, and error distribution in histograms, statistics on correlation and autocorrelation metrics based on an exhaustive simulation study for an information security model. The validation of the performance of the design nonlinear networks is further endorsed from counterpart's backpropagation schemes of Bayesian regularization and scaled conjugate gradient, based on the results of statistics in terms of mean, standard deviation, worst, and best of the convergence arcs, error distribution on heat map, inference on median with box plots, plot-matrix analysis, violin plots dynamics and computational time analysis, on exhaustive autonomous executions for solving cyber-attack model in information security.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"163 ","pages":"Article 112872"},"PeriodicalIF":8.0,"publicationDate":"2025-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145334917","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A cosmetic packaging design method based on online reviews","authors":"Zhan Gao,&nbsp;Zhenyu Li","doi":"10.1016/j.engappai.2025.112865","DOIUrl":"10.1016/j.engappai.2025.112865","url":null,"abstract":"<div><div>To address the transformation of user experience and packaging iteration in cosmetics due to the diversification of usage scenarios and demands, this study capitalizes on the advancements in artificial intelligence across user analysis, data analysis, and generative design domains, and proposes a cosmetic packaging design approach centered around online reviews. In this study, 124,879 pieces of user review data were collected from JingDong (JD), a Chinese e-commerce platform, using Python programming technology. Five topics are clustered through the application of the Latent Dirichlet Allocation (LDA) topic model. By integrating the coding of Grounded Theory, 18 demand elements within six core categories are summarized. The Kano model and the Analytic Hierarchy Process (AHP) are employed to classify and rank these demands. Notably, aspects such as strong brand recognition (M1, 0.2182), strong brand value perception (M5, 0.1129), and visually appealing and refined aesthetics (A5, 0.0983) exhibit relatively high weights. Subsequently, six lipstick packaging design schemes are developed by combining traditional software with the MidJourney generative artificial intelligence tool. Through comprehensive evaluation using the Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) method, the optimal Scheme c is identified and further optimized. This study constructs a comprehensive design strategy with user online reviews at its core, encompassing data collection, analysis, scheme design, artificial intelligence (AI)-assisted design, and evaluation. It is recommended that the application of artificial intelligence (AI)-assisted design be significantly enhanced throughout the entire design process, enabling precise and rapid generation of design schemes, streamlining the process, and shortening the development cycle.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"163 ","pages":"Article 112865"},"PeriodicalIF":8.0,"publicationDate":"2025-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145335058","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Hybrid framework of penetration resistance analysis by machine learning and finite element simulation","authors":"Xu Long,&nbsp;Irfan Ali,&nbsp;Khawaja Haseeb Maqbool,&nbsp;Muhammad Muaz Khan","doi":"10.1016/j.engappai.2025.112868","DOIUrl":"10.1016/j.engappai.2025.112868","url":null,"abstract":"<div><div>A comprehensive understanding of projectile penetration in reinforced concrete (RC) structures is essential for developing resilient defense and infrastructure systems. Such investigations provide valuable insights into the behavior of structural components under extreme loading conditions. However, accurately modeling penetration resistance remains challenging due to the complex interaction among projectile velocity, geometry, and the nonlinear behavior of concrete. To address this challenge, this study applies artificial intelligence (AI) techniques in combination with finite element (FE) simulations to enhance predictive modeling. The AI framework incorporates deep neural networks (DNN), support vector machines (SVM), and random forests (RF) for prediction and classification tasks, while Bayesian neural networks (BNN) are employed for uncertainty quantification, providing statistically reliable confidence bounds for the depth of penetration (DoP). Damage categorization is further optimized through K-means clustering, enabling clear differentiation between minor and severe damage states. The analysis is based on 540 data samples generated from a validated FE model calibrated with experimental results. The hybrid DNN–RF model achieved an R² of 0.994 for DoP prediction, while the SVM attained 99.08 % precision in damage classification and the RF achieved 98.16 % accuracy in ballistic limit prediction. The BNN yielded a 95 % confidence interval, confirming the reliability of the AI-based predictions. Among various clustering algorithms, including Density-Based Spatial Clustering of Applications with Noise, Gaussian Mixture Models, and hierarchical clustering, K-means demonstrated the best performance. The proposed AI-driven framework provides a reliable and efficient tool for rapid RC design assessment and optimization, contributing to advancements in defense, infrastructure resilience, and high-performance structural engineering.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"163 ","pages":"Article 112868"},"PeriodicalIF":8.0,"publicationDate":"2025-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145335265","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An innovative optimization strategy based on Mamba and generative adversarial networks for efficient and high-performance multimodal image fusion","authors":"Yichen Sun ,&nbsp;Mingli Dong ,&nbsp;Lianqing Zhu","doi":"10.1016/j.engappai.2025.112788","DOIUrl":"10.1016/j.engappai.2025.112788","url":null,"abstract":"<div><div>Multimodal image fusion (MIF) integrates multisource data into a single high-quality image with minimal redundancy. While deep learning has advanced MIF by improving fusion quality, convolutional neural networks (CNNs) struggle with long-range dependencies, and Transformers incur high computational costs. Additionally, preserving fine textures, suppressing noise, and achieving high efficiency remain challenges, particularly for infrared and visible image fusion (IVIF). This paper proposes MMGFuse, a novel MIF framework based on a Multi-Parallel Vision Mamba Generative Adversarial Network. MMGFuse leverages the Mamba model's efficiency and generative adversarial networks' realism, introducing a residual parallel vision Mamba (ResPViM4) module to enhance texture and detail preservation and a multi-parallel vision Mamba (MPViM) module to capture both global and local features across scales. A dual-modality image discriminator further optimizes visual quality. Experiments show that MMGFuse outperforms state-of-the-art methods in subjective visual quality and objective metrics for IVIF and medical image fusion, demonstrating its effectiveness, efficiency, and broad applicability in advancing image fusion. The codes are available at <span><span>https://github.com/sunyichen1994/MMGFuse</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"163 ","pages":"Article 112788"},"PeriodicalIF":8.0,"publicationDate":"2025-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145334886","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A fuzzy multi-criteria decision-making approach for public projects–bidders matching under heterogeneous information","authors":"Faizan Ahemad ,&nbsp;Mukesh Kumar Mehlawat ,&nbsp;Pankaj Gupta ,&nbsp;Shilpi Verma ,&nbsp;Dragan Pamucar","doi":"10.1016/j.engappai.2025.112833","DOIUrl":"10.1016/j.engappai.2025.112833","url":null,"abstract":"<div><div>This study presents an intelligent decision-support framework for addressing the Projects–Bidders Matching (PBM) problem in public procurement, designed to handle heterogeneous and uncertain information. The approach employs fuzzy set theory, through triangular fuzzy numbers, intuitionistic fuzzy sets, and linguistic evaluations, to capture vagueness, hesitancy, and imprecision in expert judgments. To determine the relative importance of criteria from project and bidder perspectives, we employ a hybrid weighting mechanism that combines deviation from a reference point with entropy-based measures to derive data-driven weights. By combining fuzzy modeling, objective weighting, and behavioral decision theory within an artificial intelligence framework, the model enhances explainability and supports data-driven decision-making under uncertainty. From an engineering perspective, the framework is applied to optimize bidder assignments in real-world Indian public procurement scenarios. A multi-objective optimization model is formulated to (i) maximize cumulative prospect values that jointly reflect individual preferences and socially influenced preferences for both bidders and projects, (ii) minimize the absolute deviation between these cumulative prospect values, ensuring fairness, transparency, and alignment and (iii) satisfy a stability constraint to ensure that no bidder–project pair has an incentive to deviate from the assigned matching. The framework’s effectiveness is demonstrated through a practical case study, and its robustness is validated through extensive sensitivity and variation analyses.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"163 ","pages":"Article 112833"},"PeriodicalIF":8.0,"publicationDate":"2025-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145334892","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Missing microseismic data imputation in tunnel monitoring using a transformer model with an integrated Gaussian mixture model","authors":"Zhihao Kuang ,&nbsp;Shaojun Li ,&nbsp;Shili Qiu ,&nbsp;Yong Huang ,&nbsp;Shuaipeng Chang","doi":"10.1016/j.engappai.2025.112771","DOIUrl":"10.1016/j.engappai.2025.112771","url":null,"abstract":"<div><div>Microseismic (MS) monitoring is essential for early warning and evaluation of structural safety in tunnel engineering. However, data loss due to environmental interference often compromises the reliability of such systems. To address this challenge, a data imputation model that integrates the Gaussian Mixture Model (GMM) with a transformer-based neural network, referred to as the GMM–Transformer model, was developed. Its performance was evaluated using real-world MS monitoring data from a deep-buried tunnel project in southwestern China. The proposed method achieves high accuracy in reconstructing missing data, with the imputed results closely matching observed values across multiple characteristic parameters. By leveraging the probabilistic nature of the Gaussian mixture distribution and Monte Carlo Dropout, the model can also quantify predictive uncertainty, yielding narrow confidence intervals that reinforce its reliability. The influence of missing data duration on the imputation quality was examined. The results imply that a missing window of approximately 3.5 h yields optimal results. A comparison between direct and indirect imputation strategies indicates that the direct approach significantly reduces reconstruction errors, from 25.73 % to 13.37 %. Additionally, benchmark comparisons with models such as random forest and long short-term memory networks show that the proposed model offers superior accuracy in recovering spatial characteristic critical to MS analysis. Overall, the GMM–Transformer model provides an effective, robust solution for dealing with data loss in MS monitoring. This work provides a forward-looking methodology and theoretical foundation for advancing artificial intelligence–based MS monitoring technologies in complex tunnel environments.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"163 ","pages":"Article 112771"},"PeriodicalIF":8.0,"publicationDate":"2025-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145334910","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Time series anomaly detection based on time–frequency domain with masking strategy and contrastive learning","authors":"Zhengkai Wang ,&nbsp;Hui Liu ,&nbsp;Longjing Kuang ,&nbsp;Xueliang Zhang ,&nbsp;Xiude Chen ,&nbsp;Junzhao Du","doi":"10.1016/j.engappai.2025.112775","DOIUrl":"10.1016/j.engappai.2025.112775","url":null,"abstract":"<div><div>Anomalies in time series often indicate underlying issues or system failures. Timely detection is critical to avoid severe consequences like system crashes and traffic accidents. Although some high-performing time series anomaly detection models already exist, several challenges remain: (1) Training Bias: Unsupervised anomaly detection models are typically trained on clean normal data. If the training data contains noise or potential anomalies, it can cause the model parameters to deviate from the ideal state during optimization, hindering accurate anomaly detection. (2) Distribution Shift: Time series exhibit periodicity and trends, and the training and testing data may have different distribution patterns. This may cause the model to incorrectly classify normal data as anomalies during testing. Therefore, we propose an anomaly detection network called TFCLNet, which utilizes a time–frequency domain masking strategy combined with contrastive learning. Since the frequency domain reveals potential periodicity and frequency variations, a dual-branch structure is adopted to simultaneously process time-domain and frequency-domain features. Additionally, we employ targeted masking strategies in both domains to reduce the impact of noise and address training bias, thereby learning the core data patterns of time series. Furthermore, unlike traditional contrastive learning strategies based on raw features, we minimize the distribution differences between the reconstructed time–frequency domain features through a contrastive objective function, mitigating the negative impact of distribution shifts in the original data on detection performance. Finally, adversarial training is incorporated to prevent overfitting. Experimental results on five real-world datasets demonstrate that TFCLNet outperforms all baseline models and achieves state-of-the-art performance.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"162 ","pages":"112775"},"PeriodicalIF":8.0,"publicationDate":"2025-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145333605","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Robust Direct-Discretized RNN for Time-Dependent Optimization Constrained by Nonlinear Equalities and Its Applications","authors":"Guangfeng Cheng;Binbin Qiu;Jinjin Guo;Yu Han","doi":"10.1109/JAS.2025.125627","DOIUrl":"https://doi.org/10.1109/JAS.2025.125627","url":null,"abstract":"In recent years, numerous recurrent neural network (RNN) models have been reported for solving time-dependent nonlinear optimization problems. However, few existing RNN models simultaneously involve nonlinear equality constraints, direct discretization, and noise suppression. This limitation presents challenges when existing models are applied to practical engineering problems. Additionally, most current discrete-time RNN models are derived from continuous-time models, which may not perform well for solving essentially discrete problems. To handle these issues, a robust direct-discretized RNN (RDD-RNN) model is proposed to efficiently realize time-dependent optimization constrained by nonlinear equalities (TDOCNE) in the presence of various time-dependent noises. Theoretical analyses are provided to reveal that the proposed RDD-RNN model possesses excellent convergence and noise-suppressing capability. Furthermore, numerical experiments and manipulator control instances are conducted and analyzed to validate the superior robustness of the proposed RDD-RNN model under various time-dependent noises, particularly quadratic polynomial noise. Eventually, small target detection experiments further demonstrate the practicality of the RDD-RNN model in image processing applications.","PeriodicalId":54230,"journal":{"name":"Ieee-Caa Journal of Automatica Sinica","volume":"12 9","pages":"1866-1877"},"PeriodicalIF":19.2,"publicationDate":"2025-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145335305","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}