检索结果-内蒙古大学图书馆

Enhancing structural health monitoring with AI-ML algorithms: a focus on crack detection and prediction

Asian Journal of Civil engineering 2025年第5期26卷 1907-1918页

作者： Bader, Ahmad Shtayat, Amir Al-Mistarehi, Bara’ Communication and Computer Department Faculty of Engineering Jadara University Irbid Jordan Department of City Planning and Design Jordan University of Science and Technology Irbid 22110 Jordan Department of Civil Engineering Jordan University of Science and Technology Irbid Jordan

SHM is a very important process in terms of the safety and durability of infrastructure. Traditional SHM often faces problems detecting minor structural defects and handling large datasets. Therefore, certain more advanced approaches are called for. The paper discussed the applications of AI and ML algorithms, such as CatBoost and the African Vultures Optimization Algorithm, for such challenges. The research is based on a unique dataset of 8,541 rows and diverse features, developing a predictive framework that enhances crack detection and forecast capabilities. The approach mainly deals with heterogeneous data using the CatBoost algorithm, given its capability for high-accuracy predictions, while AVOA optimizes feature selection, reduces the computational cost, and guarantees no loss in model performance. This methodology has resulted in a significant enhancement of the prediction accuracy, which states the importance of AI-ML integration in SHM. The key results demonstrate the effectiveness of the model in detecting structural anomalies and crack propagation to enable proactive maintenance strategies. This study’s contributions have gone toward advancing SHM with scalable and efficient AI-ML frameworks, enabling real-time monitoring for better infrastructure management. Such development might have a transforming potential to cut down on maintenance costs and enhance operational safety, thus further encouraging sustainable infrastructure systems. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2025.

关键词： African vultures optimization algorithm CatBoost Crack Detection Feature selection Machine learning Structural health monitoring

来源：评论

学校读者我要写书评

暂无评论

The Analysis of Device-to-Device (D2D) Communication System Operating in 5G Underlaying Cellular Networks on Multiple Bands over Nakagami-m Fading Channels

引用

SN computer science 2025年第5期6卷 1-15页

作者： Shatnawi, Razan A. Khodeir, Mahmoud A. Al-Mistarihi, Mamoun F. Darabkh, Khalid A. Al-Zuraiqi, Eyad T. Qasaymeh, Mahmoud M. Electrical Engineering Department Faculty of Engineering Jordan University of Science and Technology P.O. Box 3030 Irbid 22110 Jordan Computer Engineering Department School of Engineering The University of Jordan Amman 11942 Jordan Department of Telecommunications Engineering Hijjawi Faculty for Engineering Technology Yarmouk University P.O. Box 566 Irbid 21163 Jordan Computer Engineering and Communication Department Tafila Technical University Tafila 66110 Jordan

In contrast to traditional cellular connection, device-to-device (D2D) communication is a direct connection amidst adjacent mobile users that does not pass through the base station (BS) and does not rely on network infrastructure to meet low power requirements and high data rate. Cellular networks and D2D connections were combined in underlay mode to address power usage concerns. In underlay D2D communication, licensed frequency bands are utilized by both D2D and cellular connections;due to resource sharing, these bands may overlap, which lowers service quality but boosts spectral efficiency. Most existing research exclusively discuss energy efficiency (EE) in the context of single cells. This article has taken into account the entire cellular network with D2D connections on various bands, and stochastic geometric theory has been used in order to obtain closed form expressions for the probability of successful transmission (STP), the average sum rate (ASR), and EE for D2D connection underlaid cellular networks over Nakagami-m fading channels. Results of simulations are used to validate analytical conclusions. Simulation results demonstrate that higher STP, higher ASR, and lower EE all result in enhanced system performance. In addition, as the shape parameter’s value m (Nakagami-m order) increases, the EE, ASR, and STP of D2D users improve, which enhances system performance. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.

关键词： Device-to-device communication Energy efficiency Nakagami-m fading channel Stochastic geometry

来源：评论

学校读者我要写书评

暂无评论

Classifying white blood cells using combining different convolutional neural networks

引用

Multimedia Tools and Applications 2025年 1-24页

作者： Toptaş, Murat Toptaş, Buket Hanbay, Davut Software Eng. Dept. Engineering and Natural Science Faculty Bandırma Onyedi Eylül University Balıkesir Turkey Computer Eng. Dept. Engineering Faculty Inonu University Malatya44280 Turkey

White blood cells are warrior cells that protect the human body against external factors. Each of these warrior cells performs a distinct task, making every piece of information about them highly valuable in the medical field. In this article, a classification framework for the four known types of white blood cells is proposed. It is hoped that the classification of these types will contribute to the prediction of diseases such as AIDS, malaria, leukemia, and many others. In the proposed method, images of white blood cells from the Blood Cell Classification and Detection dataset were used as input to Convolutional Neural Networks. The feature vectors extracted using these Convolutional Neural Network architectures were combined into a single vector. A Minimum Redundancy Maximum Relevance algorithm was then employed to identify the most effective features within the feature vector. Experiments were conducted using these selected features, and the analysis of each experiment was reported in detail. The Support Vector Machines classifier achieved an accuracy of 98.63% in classifying white blood cell types by combining features from multiple deep learning architectures. The experimental results demonstrated that the features obtained from different layers of the Convolutional Neural Networks had varying impacts on the classification performance. © The Author(s) 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Construction of PCOS Prediction Model based on BP Neural Network 6

Construction of PCOS Prediction Model based on BP Neural Net...

引用

6th International Conference on Mobile Computing and Sustainable Informatics, ICMCSI 2025

作者： Hu, Xianqing Yadav, Amit Khan, Asif Sah, Abhishek Pratap Azam, Sami School of Computer and Software Chengdu Neusoft University Chengdu China Faculty of Science and Technology Charles Darwin University NT Australia Department of Computer Application Integral University Lucknow India School of Management Science and Engineering University of Electronic Science and Technology of China Chengdu China

ISBN: (纸本)9798331522667

Polycystic ovary syndrome (PCOS), a common endocrine-metabolic disorder affecting about 10-13% of women during reproductive age worldwide, often leads to irregular menstruation, infertility, obesity, and long-term health risks such as diabetes and cardiovascular disease. In this paper, clinical data from PCOS patients and healthy control group were collected from ten hospitals containing 37 key indicators such as age, weight, BMI, menstrual cycle, etc. BP (Back Propagation) Neural Network prediction models were constructed and compared using the collected data and the cross-validation method was used for parameter tuning. It was found that BP Neural Network performed particularly well on the test set, and both demonstrated high prediction accuracy and generalization ability which provided strong evidence for the early identification and early valuable intervention opportunities for PCOS patients. © 2025 IEEE.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Eddy Detection of Lakes in the Qinghai–Tibet Plateau Based on Optical Remote Sensing and Deep Learning

引用

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 2025年 18卷 11996-12007页

作者： Bai, Bingxin Mu, Lixia Yang, Jie Huang, Baoxiang Ma, Chunyong Chen, Ge Ocean University of China Faculty of Information Science and Engineering Qingdao266100 China Laoshan Laboratory Qingdao266237 China Ocean University of China College of Marine Technology Qingdao266100 China Qingdao University College of Computer Science and Technology Qingdao266071 China

Lake eddies are dynamic phenomena prevalent in large lake systems, playing a critical role in affecting lake physics, nutrient transport, and ecological balance. Efficient and precise detection of these features is essential for advancing scientific understanding of lake dynamics and for improving environment management strategies. Despite their importance, the study of lake eddies remains under-explored, and current detection methods are insufficient. This study addresses this gap by proposing a lake eddy identification framework based on the YOLOv7 deep learning model, utilizing Landsat 8/9 satellite imagery acquired from April 2013 to December 2023. Qinghai Lake, Selin Co, and Nam Co on the Qinghai–Tibet Plateau were selected as the study areas. The method involves the development of an image enhancement process to optimize the visibility of lake eddies in remote sensing imagery, followed by the implementation of the YOLOv7 model for detection. The framework identified 912 cyclonic eddies and 102 anticyclonic eddies on the three aforementioned lakes, providing detailed information on their spatial distribution and morphology. This approach demonstrates significant potential for advancing the study of lake hydrodynamics and energy transport substances in aquatic ecosystems. © 2008-2012 IEEE.

关键词： Aquatic ecosystems

来源：评论

学校读者我要写书评

暂无评论

ATM: Adaptive transformer model for reconstruction of remote sensing multi-temporal non-equispaced sequence imagery

引用

Journal of King Saud University - computer and Information sciences 2025年第4期37.0卷 1-13页

作者： Qi, Zhi-Peng Zhao, Zi-Yi Jin, Xing Faculty of Electrical Engineering and Computer Science Ningbo University Ningbo 315211 China

Due to the influence of the imaging characteristics of the solar orbiting satellite and atmospheric conditions, the multi-spectral observation data often have the missing of phase image, which brings difficulties to the efficient and automatic processing of regional remote sensing data. However, the observation sequence contain complex and diverse information, it is difficult for the conventional method to characterize this information. To address this issue, we perform an adaptive transformer model (ATM) for reconstruction of remote sensing multi-temporal non-equispaced sequence imagery. To realize efficient information processing by dynamically adjusting the weight distribution of the internal relations of data, the core workflow of ATM is divided into three steps: feature coding, relationship evaluation and adaptive focusing. Firstly, the input data is encoded as a sequence of vectors, each representing a local feature of the data. Then, the model generates an attention score by calculating the similarity between the query and the key, quantifying the strength of the association between the different features. Finally, a context-aware feature representation is formed by using the normalized attention score weighted aggregation vector. Our experiments, conducted utilizing unmanned aerial vehicle (UAV) datasets and the Landsat-8 datasets, demonstrate that the proposed methodology achieves promising performance in terms of both precision (RMSE decreases by about 2.5 points, and SSIM increases by about 0.05 points) and effectiveness (time cost savings of about 50 %), and the proposed model has wide application prospects in intelligent agriculture, water color inversion and vegetation phenology detection. © The Author(s) 2025.

关键词： Adaptive transformer model Missing of phase image Multi-spectral observation data Non-equispaced sequence imagery

来源：评论

学校读者我要写书评

暂无评论

DDT-Net:Deep Detail Tracking Network for Image Tampering Detection

引用

computers, Materials & Continua 2025年第5期83卷 3451-3469页

作者： Jim Wong Zhaoxiang Zang Hubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering China Three Gorges UniversityYichang443002China College of Computer and Information Technology ChinaThree Gorges UniversityYichang443002China Faculty of Information Engineering and Automation Kunming University of Science and TechnologyKunming650504China

In the field of image forensics,image tampering detection is a critical and challenging *** methods based on manually designed feature extraction typically focus on a specific type of tampering operation,which limits their effectiveness in complex scenarios involving multiple forms of *** deep learningbasedmethods offer the advantage of automatic feature learning,current approaches still require further improvements in terms of detection accuracy and computational *** address these challenges,this study applies the UNet 3+model to image tampering detection and proposes a hybrid framework,referred to as DDT-Net(Deep Detail Tracking Network),which integrates deep learning with traditional detection *** contrast to traditional additive methods,this approach innovatively applies amultiplicative fusion technique during downsampling,effectively combining the deep learning feature maps at each layer with those generated by the Bayar noise *** design enables noise residual features to guide the learning of semantic features more precisely and efficiently,thus facilitating comprehensive feature-level ***,by leveraging the complementary strengths of deep networks in capturing large-scale semantic manipulations and traditional algorithms’proficiency in detecting fine-grained local traces,the method significantly enhances the accuracy and robustness of tampered region *** with other approaches,the proposed method achieves an F1 score improvement exceeding 30% on the DEFACTO and DIS25k *** addition,it has been extensively validated on other datasets,including CASIA and *** results demonstrate that this method achieves outstanding performance across various types of image tampering detection tasks.

关键词： Image forensics image tampering detection image manipulation detection noise flow Bayar

来源：评论

学校读者我要写书评

暂无评论

Big Data Healthcare Monitoring System: Improved Recurrent Neural Network for Disease Classification with Apache Spark

引用

Sensing and Imaging 2025年第1期26卷 1-52页

作者： Prabhugouda, Amruta Asra, Syeda Department of Computer Science and Engineering Faculty of Engineering and Technology (Exclusively for Women) Sharnbasva University Karnataka Kalaburagi 585102 India Department of Computer Science and Engineering Faculty of Engineering and Technology Sharnbasva University Karnataka Kalaburagi 585102 India

Big data analytics has increasingly penetrated the medical industry due to the swift growth of Internet technology along with medical data digitalization, hospital information systems, a huge count of Electronic Health Records (EHR) and other emerging data. The big data’s potential in healthcare is mainly based on its capability of detecting patterns and turning the high volume of data into actionable knowledge for decision-makers. Considering the applications of big data analytics in the medical industry, we have introduced an improved RNN-based big data healthcare monitoring system including the following working stages. Firstly, acquired data gets pre-processed by an outlier detection process. Afterwards, Improved SMOTE (Synthetic Minority Oversampling Technique) based class imbalance processing is performed to get the balanced data. This balanced data is handled using the Spark framework, with the master node carrying out an improved Deep Fuzzy Clustering (DFC) based clustering process and the slave node handling feature extraction and an enhanced Support Vector Machine Recursive Feature Elimination (SVM-RFE) based feature selection process. To divide the data according to the patient’s condition, an enhanced Deep Autoencoder-based Fuzzy C Means Clustering (DAE-FCM) is suggested in the improved DFC. Features including statistics, enhanced entropy, and mutual information are extracted throughout the feature extraction process. Ultimately, an Improved Recurrent Neural Network (RNN) is used to classify diseases using the chosen feature from the slave node. The implementation outcomes proved that the proposed big data healthcare monitoring system can provide effective and accurate disease classification. The Improved RNN method demonstrated the highest accuracy, achieving an impressive rate of 0.945 for dataset 1 and 0.947 for dataset 2 at training data 80%, while the conventional methods acquired the least accuracy ratings. By integrating advanced techniques, the s

关键词： Bid data analytics Deep learning Healthcare data Improved DAE-FCM Improved DFC Improved entropy Improved RNN Improved SMOTE Mutual information Outlier detection

来源：评论

学校读者我要写书评

暂无评论

Load Balancing in Cloud Computing based on Ant Colony Optimization and Crow Search Algorithm 27

Load Balancing in Cloud Computing based on Ant Colony Optimi...

引用

27th International Conference on Advanced Communications Technology, ICACT 2025

作者： Alsheavi, Amar N. Wang, Xingfu Zhao, Wei Hawbani, Ammar Alhusaini, Naji Ahmed, Ibrahim Abdulrab Ismail, A.S. School of Computer Science and Technology University of Science and Technology of China Anhui Hefei230027 China School of Computer Science and Technology Anhui University of Technology Anhui China School of Computer Science Shenyang Aerospace University Shenyang110136 China School of Computer and Information Engineering Chuzhou University Anhui Chuzhou239000 China Computer Department Applied College Najran University Najran66462 Saudi Arabia College of Life Sciences and Oceanography Shenzhen University Shenzhen518060 China Faculty of Science Zagazig University Zagazig44519 Egypt

ISBN: (纸本)9791188428137

With the rapid increase in cloud services and the increasing shift toward them, balancing the cloud load has become a critical research issue. The increasing demand from customers for technology services worldwide is largely due to its direct impact on performance quality. Therefore, to provide better service quality, it is essential to consider reducing response time and cost in load balancing. This paper addresses the load-balancing problem by optimizing response time and computing cost in cloud computing systems. Specifically, we first formulated the load balancing problem in cloud computing mathematically by designing an objective function that optimizes response time and cost, including constraints to ensure task assignment to a single virtual machine (VM) and resource limits are not exceeded. To solve this optimization problem, we proposed a hybrid algorithm, ACOCSA, that combines Ant Colony Optimization (ACO) and Crow Search Algorithm (CSA). We implemented the ACOCSA algorithm, evaluated the response time, cost, and load balancing metrics, and found that our algorithm performed better in terms of response time and computational cost, and showed good fairness among VMs. © 2025 Institute of Electrical and Electronics Engineers Inc.. All rights reserved.

关键词： Ant colony optimization

来源：评论

学校读者我要写书评

暂无评论

Dysarthric speech recognition: an investigation on using depthwise separable convolutions and residual connections

引用

Neural Computing and Applications 2025年第12期37卷 7991-8005页

作者： Shahamiri, Seyed Reza Mandal, Krishnendu Sarkar, Sudeshna DeepNet Discovery Network Faculty of Engineering and Design University of Auckland Private Bag 92019 Building 405 Level 6 Room 669 3 Garfton Road Auckland1142 New Zealand Computer Science and Engineering Department Indian Institute of Technology Kharagpur India

Dysarthria is a neurological condition resulting from impairments affecting muscle control involved in speech articulation, leading to reduced intelligibility or unintelligible speech, which affects communication abilities. Although Automatic Speech Recognition (ASR) technologies hold the potential to improve the lives of people with dysarthria significantly, ASR systems designed for normal speech have shown limited effectiveness when presented with impaired speech. Consequently, researchers have focused on developing ASR systems specifically tailored for dysarthria. However, progress in this area has been gradual due to the scarcity of dysarthric speech for training and the increased variability of speech among dysarthric individuals, necessitating a larger dataset of dysarthric utterances. One potential solution to enhance the robustness of dysarthric ASR is to deepen the architecture of the acoustic model, which maps the speech signal to words or phonetic units. However, deeper architectures require more training data and pose challenges in dealing with issues such as the vanishing gradient problem and representational bottlenecks in deep learning models. In this study, we expanded on our previous findings and investigated the applications of Depthwise Separable Convolution neurons and the inclusion of Residual Connections to propose a deep dysarthric acoustic model, tackling both vanishing gradients and representational bottleneck issues in dysarthric ASR. Multiple speaker-adaptive dysarthric ASRs were trained and evaluated for 15 UA-Speech dysarthric subjects, then benchmarked against the state-of-the-art and our previous dysarthric ASRs. Our proposed architectures have delivered up to 22.58% word recognition rate (WRR) improvements over the reference models. We observed an average of 10.81% better WRRs over the base traditional dysarthric ASR for all speakers. Likewise, the proposed acoustic model outperformed the state-of-the-art Transformer-based dysarthric

关键词： Convolution

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：