检索结果-内蒙古大学图书馆

Topological search and gradient descent boosted Runge-Kutta optimiser with application to engineering design and feature selection

引用

CAAI Transactions on intelligence Technology 2025年第2期10卷 557-614页

作者： Jinge Shi Yi Chen Ali Asghar Heidari Zhennao Cai Huiling Chen Guoxi Liang Department of Computer Science and Artificial Intelligence Wenzhou UniversityWenzhouChina School of Surveying and Geospatial Engineering College of EngineeringUniversity of TehranTehranIran Department of Artificial Intelligence Wenzhou PolytechnicWenzhouChina

The Runge-Kutta optimiser(RUN)algorithm,renowned for its powerful optimisation capabilities,faces challenges in dealing with increasing complexity in real-world ***,it shows deficiencies in terms of limited local exploration capabilities and less precise ***,this research aims to integrate the topological search(TS)mechanism with the gradient search rule(GSR)into the framework of RUN,introducing an enhanced algorithm called TGRUN to improve the performance of the original *** TS mechanism employs a circular topological scheme to conduct a thorough exploration of solution regions surrounding each solution,enabling a careful examination of valuable solution areas and enhancing the algorithm’s effectiveness in local *** prevent the algorithm from becoming trapped in local optima,the GSR also integrates gradient descent principles to direct the algorithm in a wider investigation of the global solution *** study conducted a serious of experiments on the IEEE CEC2017 comprehensive benchmark function to assess the enhanced effectiveness of ***,the evaluation includes real-world engineering design and feature selection problems serving as an additional test for assessing the optimisation capabilities of the *** validation outcomes indicate a significant improvement in the optimisation capabilities and solution accuracy of TGRUN.

关键词： engineering design gradient search rule metaheuristic algorithm Runge-Kutta optimizer topological search

来源：评论

学校读者我要写书评

暂无评论

Improved Runge Kutta Optimization Using Compound Mutation Strategy in Reinforcement Learning Decision Making for Feature Selection

引用

Journal of Bionic engineering 2024年第5期21卷 2460-2496页

作者： Jinpeng Huang Yi Chen Ali Asghar Heidari Lei Liu Huiling Chen Guoxi Liang Department of Computer Science and Artificial Intelligence Wenzhou UniversityWenzhou325035China School of Surveying and Geospatial Engineering College of EngineeringUniversity of TehranTehran***Iran College of Computer Science Sichuan UniversityChengdu610065China Department of Artificial Intelligence Wenzhou PolytechnicWenzhou325035China

Runge Kutta Optimization(RUN)is a widely utilized metaheuristic ***,it suffers from these issues:the imbalance between exploration and exploitation and the tendency to fall into local optima when it solves real-world opti-mization *** address these challenges,this study aims to endow each individual in the population with a certain level of intelligence,allowing them to make autonomous decisions about their next optimization *** incorporating Reinforcement Learning(RL)and the Composite Mutation Strategy(CMS),each individual in the population goes through additional self-improvement steps after completing the original algorithmic phases,referred to as *** is,each individual in the RUN population is trained intelligently using RL to independently choose three different differentiation strategies in CMS when solving different *** validate the competitiveness of RLRUN,comprehensive empirical tests were conducted using the IEEE CEC 2017 benchmark *** comparative experiments with 13 conventional algorithms and 10 advanced algorithms were *** experimental results demonstrated that RLRUN excels in convergence accuracy and speed,surpassing even some champion ***,this study introduced a binary version of RLRUN,named bRLRUN,which was employed for the feature selection *** 24 high-dimensional datasets encompassing UCI datasets and SBCB machine learning library microarray datasets,bRLRUN occupies the top position in classification accuracy and the number of selected feature subsets compared to some *** conclusion,the proposed algorithm demonstrated that it exhibits a strong competitive advantage in high-dimensional feature selection for complex datasets.

关键词： Runge Kutta optimization Metaheuristic algorithm Feature selection Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Learner Phase of Partial Reinforcement Optimizer with Nelder-Mead Simplex for Parameter Extraction of Photovoltaic Models

引用

Journal of Bionic engineering 2024年第6期21卷 3041-3075页

作者： Jinpeng Huang Zhennao Cai Ali Asghar Heidari Lei Liu Huiling Chen Guoxi Liang Department of Computer Science and Artificial Intelligence Wenzhou UniversityWenzhou325035China School of Surveying and Geospatial Engineering College of EngineeringUniversity of TehranTehran***Iran College of Computer Science Sichuan UniversityChengdu610065SichuanChina Department of Artificial Intelligence Wenzhou PolytechnicWenzhou325035China

This paper proposes an improved version of the Partial Reinforcement Optimizer(PRO),termed *** LNPRO has undergone a learner phase,which allows for further communication of information among the PRO population,changing the state of the PRO in terms of ***,the Nelder-Mead simplex is used to optimize the best agent in the population,accelerating the convergence speed and improving the accuracy of the PRO *** comparing LNPRO with nine advanced algorithms in the IEEE CEC 2022 benchmark function,the convergence accuracy of the LNPRO has been *** accuracy and stability of simulated data and real data in the parameter extraction of PV systems are *** to the PRO,the precision and stability of LNPRO have indeed been enhanced in four types of photovoltaic components,and it is also superior to other excellent *** further verify the parameter extraction problem of LNPRO in complex environments,LNPRO has been applied to three types of manufacturer data,demonstrating excellent results under varying irradiation and *** summary,LNPRO holds immense potential in solving the parameter extraction problems in PV systems.

关键词： Partial reinforcement optimizer Learner phase Nelder-Mead simplex algorithm Parameter extraction

来源：评论

学校读者我要写书评

暂无评论

Enhance Dehazed Images Rapidly Without Losing Restoration Accuracy

引用

IEEE Access 2024年 12卷 40795-40808页

作者： Liu, Ping-Juei National Chin-Yi University of Technology Department of Artificial Intelligence and Computer Engineering Taichung411030 Taiwan

We proposed a novel image-enhancing framework to ensure consolidated restoration accuracy when remedying the visual quality of dehazed images, such as over-saturation, color deviation, or luminance issues. Conventionally, the dehazing process was usually considered image restoration;however, enhancement methods only aimed to improve dehazed images' contrast, brightness, and detail. Therefore, the conventional enhancing framework tended to degrade consolidated restoration accuracy due to focusing on low-level image features rather than controlling errors associated with the dehazing process. Our method aimed to improve the consolidated restoration accuracy of the dehazing-and-enhancing process. In experiments, the proposed framework improved the visual quality and preserved restoration accuracy (despite enhancements) with high computational efficiency and resolved quality issues generated by conventional dehazing algorithms. Moreover, the minimal time complexity of the proposed framework is O(n), ensuring practical applicability when implemented in conjunction with state-of-the-art dehazing algorithms. © 2013 IEEE.

关键词： Probability density function

来源：评论

学校读者我要写书评

暂无评论

A Framework for Enhancing Privacy and Anonymity in Blockchain-Enabled IoT Devices

引用

computers, Materials & Continua 2024年第3期78卷 4263-4282页

作者： Muhammad Saad Muhammad Raheel Bhutta Jongik Kim Tae-Sun Chung Department of Artificial Intelligence Ajou UniversitySuwon443-749Korea Department of Electrical and Computer Engineering University of Utah Asia CampusIncheon21985Korea Department of Artificial Intelligence Chungnam National UniversityDaejeon34134Korea

With the increase in IoT(Internet of Things)devices comes an inherent challenge of *** the world today,privacy is the prime concern of every *** one’s privacy and keeping anonymity throughout the system is a desired functionality that does not come without inevitable trade-offs like scalability and increased complexity and is always exceedingly difficult to *** challenge is keeping confidentiality and continuing to make the person innominate throughout the *** address this,we present our proposed architecture where we manage IoT devices using blockchain *** proposed architecture works on and off blockchain integrated with the closed-circuit television(CCTV)security camera fixed at the rental *** this framework,the CCTV security camera feed is redirected towards the owner and renter based on the smart contract *** entity(owner or renter)can see the CCTV security camera feed at one *** is no third-party dependence except for the CCTV security camera deployment *** contributions include the proposition of framework architecture,a novel smart contract algorithm,and the modification to the ring signatures leveraging an existing cryptographic *** are made based on different systems’security and key management *** an empirical study,our proposed algorithm performed better in key generation,proof generation,and verification *** comparing similar existing schemes,we have shown the proposed architectures’*** now,we have developed this system for a specific area in the real ***,this system is scalable and applicable to other areas like healthcare monitoring systems,which is part of our future work.

关键词： Privacy anonymity blockchain IoT smart contracts

来源：评论

学校读者我要写书评

暂无评论

Deep learning-based face detection and recognition on drones

引用

Journal of Ambient intelligence and Humanized Computing 2024年第1期15卷 373-387页

作者： Rostami, Mohsen Farajollahi, Amirhamzeh Parvin, Hashem Department of Engineering University of Imam Ali Tehran Iran Department of Artificial Intelligence Faculty of Computer Engineering University of Isfahan Isfahan Iran

Unmanned aerial vehicles as known as drones, are aircraft that can comfortably search locations which are excessively dangerous or difficult for humans and take data from bird's-eye view. Enabling unmanned aerial vehicles to detect and recognize humans on the ground is essential for various applications, such as remote monitoring, people search, and surveillance. The current face detection and recognition models are able to detect or recognize faces on unmanned aerial vehicles using various limits in height, angle and distance, mainly where drones take images from high altitude or long distance. In the present paper, we proposed a novel face detection and recognition model on drones for improving the performance of face recognition when query images are taken from high altitudes or long distances that do not show much facial information of the humans. Moreover, we aim to employ deep neural network to perform these tasks and reach an enhanced top performance. Experimental evaluation of the proposed framework compared to state-of-the-art models over the DroneFace dataset demonstrates that our method can attain competitive accuracy on both the recognition and detection protocols. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2022.

关键词： Drones

来源：评论

学校读者我要写书评

暂无评论

Advancing Alzheimer’s detection: integrative approaches in MRI analysis with traditional and deep learning models

引用

Neural Computing and Applications 2025年第14期37卷 8527-8546页

作者： Vanaja, T. Shanmugavadivel, Kogilavani Subramanian, Malliga Kanimozhiselvi, C.S. Department of Artificial Intelligence Kongu Engineering College Tamil Nadu Erode India Department of Computer Science and Engineering Kongu Engineering College Tamil Nadu Erode India

Alzheimer’s dementia (AD) poses a significant global health challenge, characterized by progressive cognitive decline, memory impairment, and behavioral changes. The critical need for early detection to enable timely intervention and personalized care is emphasized by the current lack of effective treatments. This study aims to develop precise diagnostic models for AD by employing machine learning and a customized deep-convolutional neural network (cDCNN) with three convolution layers, utilizing Magnetic Resonance Imaging (MRI) data. Methods involve analyzing two distinct datasets—Alzheimer’s Disease Neuroimaging Initiative (ADNI) and Kaggle—to explore diverse cohorts and imaging features associated with AD pathology. To address class imbalance, the Synthetic Minority Over-sampling Technique (SMOTE) is applied across both datasets. A range of traditional machine learning methods, including support vector machine, k-nearest neighbor, random forest, decision trees, and XGBoost classifier, are evaluated alongside the cDCNN model, which leverages key MRI biomarkers of AD for both datasets. Results show the cDCNN model achieved a specific accuracy of 87% on the ADNI dataset, despite challenges in converting ADNI’s Digital Imaging and Communications in Medicine (DICOM) files to JPEG, impacting image clarity. Conclusions suggest that this research provides critical diagnostic tools for clinicians, offering insights into AD pathology and contributing to the alleviation of AD’s societal impact. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

An Overview of LoRa Localization Technologies

引用

computers, Materials & Continua 2025年第2期82卷 1645-1680页

作者： Huajiang Ruan Panjun Sun Yuanyuan Dong Hamid Tahaei Zhaoxi Fang Institute of Artificial Intelligence Shaoxing UniversityShaoxing312000China Department of Computer Science and Engineering School of Mechanical and Electrical EngineeringShaoxing UniversityShaoxing312000China

Traditional Global Positioning System(GPS)technology,with its high power consumption and limited perfor-mance in obstructed environments,is unsuitable for many Internet of Things(IoT)*** paper explores LoRa as an alternative localization technology,leveraging its low power consumption,robust indoor penetration,and extensive coverage area,which render it highly suitable for diverse IoT *** comprehensively review several LoRa-based localization techniques,including time of arrival(ToA),time difference of arrival(TDoA),round trip time(RTT),received signal strength indicator(RSSI),and fingerprinting *** this review,we evaluate the strengths and limitations of each technique and investigate hybrid models to potentially improve positioning *** studies in smart cities,agriculture,and logistics exemplify the versatility of LoRa for indoor and outdoor *** findings demonstrate that LoRa technology not only overcomes the limitations of GPS regarding power consumption and coverage but also enhances the scalability and efficiency of IoT deployments in complex environments.

关键词： LoRa localization IoT TDoA RSSI application hybrid models

来源：评论

学校读者我要写书评

暂无评论

Block Incremental Dense Tucker Decomposition with Application to Spatial and Temporal Analysis of Air Quality Data

引用

computer Modeling in engineering & Sciences 2024年第4期139卷 319-336页

作者： SangSeok Lee HaeWon Moon Lee Sael Department of Artificial Intelligence Ajou UniversitySuwon-si16499Korea Department of Software and Computer Engineering Ajou UniversitySuwon-si16499Korea

How can we efficiently store and mine dynamically generated dense tensors for modeling the behavior of multidimensional dynamic data?Much of the multidimensional dynamic data in the real world is generated in the form of time-growing *** example,air quality tensor data consists of multiple sensory values gathered from wide locations for a long *** data,accumulated over time,is redundant and consumes a lot ofmemory in its raw *** need a way to efficiently store dynamically generated tensor data that increase over time and to model their behavior on demand between arbitrary time *** this end,we propose a Block IncrementalDense Tucker Decomposition(BID-Tucker)method for efficient storage and on-demand modeling ofmultidimensional spatiotemporal *** that tensors come in unit blocks where only the time domain changes,our proposed BID-Tucker first slices the blocks into matrices and decomposes them via singular value decomposition(SVD).The SVDs of the time×space sliced matrices are stored instead of the raw tensor blocks to save *** modeling from data is required at particular time blocks,the SVDs of corresponding time blocks are retrieved and incremented to be used for Tucker *** factor matrices and core tensor of the decomposed results can then be used for further data *** compared our proposed BID-Tucker with D-Tucker,which our method extends,and vanilla Tucker *** show that our BID-Tucker is faster than both D-Tucker and vanilla Tucker decomposition and uses less memory for storage with a comparable reconstruction *** applied our proposed BID-Tucker to model the spatial and temporal trends of air quality data collected in South Korea from 2018 to *** were able to model the spatial and temporal air quality *** were also able to verify unusual events,such as chronic ozone alerts and large fire events.

关键词： Dynamic decomposition tucker tensor tensor factorization spatiotemporal data tensor analysis air quality

来源：评论

学校读者我要写书评

暂无评论

Deep learning-based method for detection of copy-move forgery in videos

引用

Neural Computing and Applications 2025年第14期37卷 8451-8464页

作者： Elbarougy, Reda Abdelfatah, Osama Behery, G.M. El-Badry, Noha M. Department of Artificial Intelligence and Data Science College of Computer Science and Engineering University of Ha’il Ha’il81481 Saudi Arabia Department of Information Technology Faculty of Computer and Artificial Intelligence Damietta University New Damietta Kafr Saad Egypt Department of Computer Science Faculty of Computer and Artificial Intelligence Damietta University New Damietta Kafr Saad Egypt Department of Mathematics Faculty of Science Damietta University New Damietta Kafr Saad Egypt

Video forgery is one of the most serious problems affecting the credibility and reliability of video content. Therefore, detecting video forgery presents a major challenge for researchers due to the diversity of forgery types, the modernity of the programs used in forgery operations, and the abundance of information and content present in videos. The seriousness of this issue arises from the widespread use of videos in vital fields that require very high accuracy with no room for doubt or error, such as courtrooms, journalism, and others. Copy-move forgery is one of the most common and dangerous types of video forgery because of the difficulty in identifying it with the naked eye and the great diversity of forgery techniques involved in this particular type. One of the biggest challenges researchers have faced in the past is the complexity of the steps required to detect forgery in videos, leading to significant computational complexity and time consumption. The proposed method also aims to achieve better results than previous methods while reducing computational operations. Ultimately, forgery is detected with great efficiency. Compared to previous methods used for detecting copy-move forgery in the Rewind dataset, the proposed method achieves the highest F1, reaching 0.86, with a significant difference of 0.13 compared to the best result of the previous methods. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：