Engineering optimization problems are often characterized by high dimensionality,constraints,and complex,multimodal landscapes.Traditional deterministic methods frequently struggle under such conditions,prompting incr...Engineering optimization problems are often characterized by high dimensionality,constraints,and complex,multimodal landscapes.Traditional deterministic methods frequently struggle under such conditions,prompting increased interest in swarm intelligence algorithms.Among these,the Cuckoo Search(CS)algorithm stands out for its promising global search capabilities.However,it often suffers from premature convergence when tackling complex problems.To address this limitation,this paper proposes a Grouped Dynamic Adaptive CS(GDACS)algorithm.Theenhancements incorporated intoGDACS can be summarized into two key aspects.Firstly,a chaotic map is employed to generate initial solutions,leveraging the inherent randomness of chaotic sequences to ensure a more uniform distribution across the search space and enhance population diversity from the outset.Secondly,Cauchy and Levy strategies replace the standard CS population update.This strategy involves evaluating the fitness of candidate solutions to dynamically group the population based on performance.Different step-size adaptation strategies are then applied to distinct groups,enabling an adaptive search mechanism that balances exploration and exploitation.Experiments were conducted on six benchmark functions and four constrained engineering design problems,and the results indicate that the proposed GDACS achieves good search efficiency and produces more accurate optimization results compared with other state-of-the-art algorithms.展开更多
In this paper,we propose a new full-Newton step feasible interior-point algorithm for the special weighted linear complementarity problems.The proposed algorithm employs the technique of algebraic equivalent transform...In this paper,we propose a new full-Newton step feasible interior-point algorithm for the special weighted linear complementarity problems.The proposed algorithm employs the technique of algebraic equivalent transformation to derive the search direction.It is shown that the proximity measure reduces quadratically at each iteration.Moreover,the iteration bound of the algorithm is as good as the best-known polynomial complexity for these types of problems.Furthermore,numerical results are presented to show the efficiency of the proposed algorithm.展开更多
Low-light image enhancement aims to improve the visibility of severely degraded images captured under insufficient illumination,alleviating the adverse effects of illumination degradation on image quality.Traditional ...Low-light image enhancement aims to improve the visibility of severely degraded images captured under insufficient illumination,alleviating the adverse effects of illumination degradation on image quality.Traditional Retinex-based approaches,inspired by human visual perception of brightness and color,decompose an image into illumination and reflectance components to restore fine details.However,their limited capacity for handling noise and complex lighting conditions often leads to distortions and artifacts in the enhanced results,particularly under extreme low-light scenarios.Although deep learning methods built upon Retinex theory have recently advanced the field,most still suffer frominsufficient interpretability and sub-optimal enhancement performance.This paper presents RetinexWT,a novel framework that tightly integrates classical Retinex theory with modern deep learning.Following Retinex principles,RetinexWT employs wavelet transforms to estimate illumination maps for brightness adjustment.A detail-recovery module that synergistically combines Vision Transformer(ViT)and wavelet transforms is then introduced to guide the restoration of lost details,thereby improving overall image quality.Within the framework,wavelet decomposition splits input features into high-frequency and low-frequency components,enabling scale-specific processing of global illumination/color cues and fine textures.Furthermore,a gating mechanism selectively fuses down-sampled and up-sampled features,while an attention-based fusion strategy enhances model interpretability.Extensive experiments on the LOL dataset demonstrate that RetinexWT surpasses existing Retinex-oriented deeplearning methods,achieving an average Peak Signal-to-Noise Ratio(PSNR)improvement of 0.22 dB over the current StateOfTheArt(SOTA),thereby confirming its superiority in low-light image enhancement.Code is available at http://gffzz188fe103f8f1460asc06f0vuwfxo06nvf.ffgz.tsg.suse.edu.cn/CHEN-hJ516/RetinexWT(accessed on 14 October 2025).展开更多
In eld seismic data acquisition,seismic traces are often aected by substantial data gaps and strong noise interference due to environmental and instrumental factors,thus degrading the resolution and signalto-noise rat...In eld seismic data acquisition,seismic traces are often aected by substantial data gaps and strong noise interference due to environmental and instrumental factors,thus degrading the resolution and signalto-noise ratio(SNR)of the seismic profiles.Effective seismic data reconstruction and noise suppression techniques are therefore essential to recover missing signals and improve data quality.In this study,a fast projection onto convex sets(FPOCS)algorithm is proposed by incorporating an inertial parameter that involves a linear combination of the two preceding iterations based on the traditional projection onto convex sets(POCS)algorithm.Then,a weighting factor is introduced to achieve simultaneous data reconstruction and noise suppression using the weighted fast projection onto convex sets(WFPOCS)algorithm.To further suppress residual random noise in the updated solution,an optimization strategy is adopted by swapping the order of the iterative hard thresholding operator and the projection operator.The nal algorithm,termed the improved weighted fast projection onto convex sets(IWFPOCS),achieves high-efciency reconstruction and effective noise suppression.Compared with WFPOCS,the proposed method maintains fast reconstruction speed while demonstrating superior denoising performance on irregularly missing and noisy datasets.Field data experiments conrm that the proposed method signicantly improves the SNR and resolution of seismic data,oering strong practical potential for subsequent processing and interpretation.展开更多
The rapid proliferation of the Internet of Things(IoT)and cyber-physical systems(CPS)within critical infrastructure sectors has significantly expanded the attack surface for advanced and stealthy cyber threats.Since t...The rapid proliferation of the Internet of Things(IoT)and cyber-physical systems(CPS)within critical infrastructure sectors has significantly expanded the attack surface for advanced and stealthy cyber threats.Since these systems increasingly rely on real-time data exchange and autonomous control,developing intelligent,scalable,and adaptive anomaly detection mechanisms has become a pressing requirement.This paper proposes a novel hybrid framework,evolutionary-transformer-long short-term memory(Evo-Transformer-LSTM),that integrates the temporal modeling capability of LSTM networks,the global attention mechanism of Transformer encoders,and the optimization power of the improved chimp optimization algorithm(IChOA)for hyper-parameter tuning.In the proposed architecture,the Transformer encoder extracts high-level contextual patterns from traffic sequences,while the LSTM component captures local temporal dependencies.The framework is rigorously evaluated on four benchmark datasets from the Canadian Institute for Cybersecurity(CIC):CIC-IDS-2017,CSE-CIC-IDS-2018,CIC-IoT-DIAD(2024),and CIC-IoV(2024).Comparative experiments are conducted against several state-of-the-art baselines,including transformer,LSTM,bidirectional encoder representations from transformers(BERT),deep reinforcement learning(DRL),convolutional neural network(CNN),k-nearest neighbors(KNN),and random forest(RF)classifiers.Results show that the proposed Evo-Transformer-LSTM achieves up to 98.25%accuracy,an F1-score of 97.91%,and an area under the curve(AUC)of 99.36%on CIC-IDS 2017,while maintaining above 96%accuracy and 98%AUC even on the more challenging CIC-IoV 2024 dataset,consistently surpassing all baseline models.In addition,statistical significance tests confirm the superiority of the proposed approach.In conclusion,Evo-Transformer-LSTM offers a unified,scalable,and robust solution for anomaly detection in modern IoT and CPS infrastructures,with potential for real-world deployment in security-sensitive domains.展开更多
基金supported in part by the Ministry of Higher Education Malaysia(MOHE)through Fundamental Research Grant Scheme(FRGS)Ref:FRGS/1/2024/ICT02/UTM/02/10,Vot.No:R.J130000.7828.5F748the Scientific Research Project of Education Department of Hunan Province(Nos.22B1046 and 24A0771).
摘要Engineering optimization problems are often characterized by high dimensionality,constraints,and complex,multimodal landscapes.Traditional deterministic methods frequently struggle under such conditions,prompting increased interest in swarm intelligence algorithms.Among these,the Cuckoo Search(CS)algorithm stands out for its promising global search capabilities.However,it often suffers from premature convergence when tackling complex problems.To address this limitation,this paper proposes a Grouped Dynamic Adaptive CS(GDACS)algorithm.Theenhancements incorporated intoGDACS can be summarized into two key aspects.Firstly,a chaotic map is employed to generate initial solutions,leveraging the inherent randomness of chaotic sequences to ensure a more uniform distribution across the search space and enhance population diversity from the outset.Secondly,Cauchy and Levy strategies replace the standard CS population update.This strategy involves evaluating the fitness of candidate solutions to dynamically group the population based on performance.Different step-size adaptation strategies are then applied to distinct groups,enabling an adaptive search mechanism that balances exploration and exploitation.Experiments were conducted on six benchmark functions and four constrained engineering design problems,and the results indicate that the proposed GDACS achieves good search efficiency and produces more accurate optimization results compared with other state-of-the-art algorithms.
基金Supported by the Optimisation Theory and Algorithm Research Team(Grant No.23kytdzd004)University Science Research Project of Anhui Province(Grant No.2024AH050631)the General Programs for Young Teacher Cultivation of Educational Commission of Anhui Province(Grant No.YQYB2023090).
摘要In this paper,we propose a new full-Newton step feasible interior-point algorithm for the special weighted linear complementarity problems.The proposed algorithm employs the technique of algebraic equivalent transformation to derive the search direction.It is shown that the proximity measure reduces quadratically at each iteration.Moreover,the iteration bound of the algorithm is as good as the best-known polynomial complexity for these types of problems.Furthermore,numerical results are presented to show the efficiency of the proposed algorithm.
基金supported in part by the National Natural Science Foundation of China[Grant number 62471075]the Major Science and Technology Project Grant of the Chongqing Municipal Education Commission[Grant number KJZD-M202301901].
摘要Low-light image enhancement aims to improve the visibility of severely degraded images captured under insufficient illumination,alleviating the adverse effects of illumination degradation on image quality.Traditional Retinex-based approaches,inspired by human visual perception of brightness and color,decompose an image into illumination and reflectance components to restore fine details.However,their limited capacity for handling noise and complex lighting conditions often leads to distortions and artifacts in the enhanced results,particularly under extreme low-light scenarios.Although deep learning methods built upon Retinex theory have recently advanced the field,most still suffer frominsufficient interpretability and sub-optimal enhancement performance.This paper presents RetinexWT,a novel framework that tightly integrates classical Retinex theory with modern deep learning.Following Retinex principles,RetinexWT employs wavelet transforms to estimate illumination maps for brightness adjustment.A detail-recovery module that synergistically combines Vision Transformer(ViT)and wavelet transforms is then introduced to guide the restoration of lost details,thereby improving overall image quality.Within the framework,wavelet decomposition splits input features into high-frequency and low-frequency components,enabling scale-specific processing of global illumination/color cues and fine textures.Furthermore,a gating mechanism selectively fuses down-sampled and up-sampled features,while an attention-based fusion strategy enhances model interpretability.Extensive experiments on the LOL dataset demonstrate that RetinexWT surpasses existing Retinex-oriented deeplearning methods,achieving an average Peak Signal-to-Noise Ratio(PSNR)improvement of 0.22 dB over the current StateOfTheArt(SOTA),thereby confirming its superiority in low-light image enhancement.Code is available at http://gffzz188fe103f8f1460asc06f0vuwfxo06nvf.ffgz.tsg.suse.edu.cn/CHEN-hJ516/RetinexWT(accessed on 14 October 2025).
基金supported in part by the Foundation of National Key Laboratory of Uranium Resources Exploration-Mining and Nuclear Remote Sensing under Grant 2024QZ-TD-13in part by the National Natural Science Foundation of China under Grant 42564006+1 种基金in part by the Natural Science Foundation of Jiangxi Province under Grant 20242BAB26051in part by the Open Fund of SINOPEC Key Laboratory of Geophysics,and in part by support the plan of Ganpo Juncai under Grant 20243BCE51012.
摘要In eld seismic data acquisition,seismic traces are often aected by substantial data gaps and strong noise interference due to environmental and instrumental factors,thus degrading the resolution and signalto-noise ratio(SNR)of the seismic profiles.Effective seismic data reconstruction and noise suppression techniques are therefore essential to recover missing signals and improve data quality.In this study,a fast projection onto convex sets(FPOCS)algorithm is proposed by incorporating an inertial parameter that involves a linear combination of the two preceding iterations based on the traditional projection onto convex sets(POCS)algorithm.Then,a weighting factor is introduced to achieve simultaneous data reconstruction and noise suppression using the weighted fast projection onto convex sets(WFPOCS)algorithm.To further suppress residual random noise in the updated solution,an optimization strategy is adopted by swapping the order of the iterative hard thresholding operator and the projection operator.The nal algorithm,termed the improved weighted fast projection onto convex sets(IWFPOCS),achieves high-efciency reconstruction and effective noise suppression.Compared with WFPOCS,the proposed method maintains fast reconstruction speed while demonstrating superior denoising performance on irregularly missing and noisy datasets.Field data experiments conrm that the proposed method signicantly improves the SNR and resolution of seismic data,oering strong practical potential for subsequent processing and interpretation.
摘要The rapid proliferation of the Internet of Things(IoT)and cyber-physical systems(CPS)within critical infrastructure sectors has significantly expanded the attack surface for advanced and stealthy cyber threats.Since these systems increasingly rely on real-time data exchange and autonomous control,developing intelligent,scalable,and adaptive anomaly detection mechanisms has become a pressing requirement.This paper proposes a novel hybrid framework,evolutionary-transformer-long short-term memory(Evo-Transformer-LSTM),that integrates the temporal modeling capability of LSTM networks,the global attention mechanism of Transformer encoders,and the optimization power of the improved chimp optimization algorithm(IChOA)for hyper-parameter tuning.In the proposed architecture,the Transformer encoder extracts high-level contextual patterns from traffic sequences,while the LSTM component captures local temporal dependencies.The framework is rigorously evaluated on four benchmark datasets from the Canadian Institute for Cybersecurity(CIC):CIC-IDS-2017,CSE-CIC-IDS-2018,CIC-IoT-DIAD(2024),and CIC-IoV(2024).Comparative experiments are conducted against several state-of-the-art baselines,including transformer,LSTM,bidirectional encoder representations from transformers(BERT),deep reinforcement learning(DRL),convolutional neural network(CNN),k-nearest neighbors(KNN),and random forest(RF)classifiers.Results show that the proposed Evo-Transformer-LSTM achieves up to 98.25%accuracy,an F1-score of 97.91%,and an area under the curve(AUC)of 99.36%on CIC-IDS 2017,while maintaining above 96%accuracy and 98%AUC even on the more challenging CIC-IoV 2024 dataset,consistently surpassing all baseline models.In addition,statistical significance tests confirm the superiority of the proposed approach.In conclusion,Evo-Transformer-LSTM offers a unified,scalable,and robust solution for anomaly detection in modern IoT and CPS infrastructures,with potential for real-world deployment in security-sensitive domains.