检索结果-内蒙古大学图书馆

Application of data twinning based on deep time series model in smart city traffic flow prediction

Discover Internet of Things 2025年第1期5卷 1-21页

作者： Gao, Li Faculty of Computer and Software Engineering Huaiyin Institute of Technology Huai’an223003 China

This paper introduces an intelligent traffic flow prediction system that combines data twinning and deep learning, aiming to improve the prediction accuracy and model adaptability by integrating grey prediction model (GM(1,1)), long-short-term memory network (LSTM) and particle swarm optimization (PSO). The system construction starts from physical layer data acquisition, deals with missing data through smoothing and interpolation, and applies the GM(1,1) model to construct a digital twin layer for preliminary prediction. The information fusion mechanism combines real-time data and forecast data to optimize the model inputs, and the PSO algorithm is used to optimize the model parameters. The LSTM model is used as a deep time-series model, and combined with the output of the data twin prediction to make a comprehensive prediction, and at the same time, feature engineering is utilized to enhance the model performance. The system implements dynamic feedback adjustment to optimize the model in real time according to the prediction error. Empirical analysis shows that the LSTM model with integrated data twin (LSTM + DT) outperforms the traditional model in prediction accuracy, stability, and generalization ability, especially in peak hours, holidays, and emergency response, and provides a reliable solution for intelligent traffic management despite the long training time. © The Author(s) 2025.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Impact of optimizers functions on detection of Melanoma using transfer learning architectures

引用

Multimedia Tools and Applications 2025年第14期84卷 13787-13807页

作者： Kılıçarslan, Serhat Aydın, Hatice Aktas Adem, Kemal Yılmaz, Esra Kavalcı Software Engineering Bandırma Onyedi Eylul University Balıkesir Turkey Computer Engineering Sivas University of Science and Technology Sivas Turkey Computer Engineering Sivas Cumhuriyet University Sivas Turkey

Early diagnosis-treatment of melanoma is very important because of its dangerous nature and rapid spread. When diagnosed correctly and early, the recovery rate of patients increases significantly. Physical methods are not sufficient for diagnosis and classification. The aim of this study is to use a hybrid method that combines different deep learning methods in the classification of melanoma and to investigate the effect of optimizer methods used in deep learning methods on classification performance. In the study, Melanoma detection was carried out from the skin lesions image through a simulation created with the deep learning architectures DenseNet, InceptionV3, ResNet50, InceptionResNetV2 and MobileNet and seven optimizers: SGD, Adam, RmsProp, AdaDelta, AdaGrad, Adamax and Nadam. The results of the study show that SGD has better and more stable performance in terms of convergence rate, training speed and performance than other optimizers. In addition, the momentum parameter added to the structure of the SGD optimizer reduces the oscillation and training time compared to other functions. It was observed that the best melanoma detection among the combined methods was achieved using the DenseNet model and SGD optimizer with a test accuracy of 0.949, test sensitivity 0.9403, and test F score 0.9492. © The Author(s) 2024.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

A mathematical multi-objective optimization model and metaheuristic algorithm for effective advertising in the social internet of things

引用

Neural Computing and Applications 2025年第6期37卷 4797-4821页

作者： Molaei, Reza Rahsepar Fard, Kheirollah Bouyer, Asgarali Department of Computer Engineering and Information Technology University of Qom Qom Iran Department of Software Engineering Azarbaijan Shahid Madani University Tabriz Iran Department of Software Engineering Faculty of Engineering and Natural Science Istinye University Istanbul Turkey

The integration of social networks with the Internet of Things (IoT) has been explored in recent research, giving rise to the Social Internet of Things (SIoT). One promising application of SIoT is viral marketing, which plays a critical role in modern advertising. However, current solutions proposed for viral marketing are often impractical in real-world scenarios, and there has been limited investigation into viral marketing within the context of SIoT. To address this issue, we present a multi-objective integer linear programming model named ILP-ASoT that facilitates advertising through SIoT. Our approach introduces two realistic parameters: the node sociality rate and the node appropriateness rate as an advertising destination. By balancing the cost of selecting a seed set with the number of users who receive the advertisement, our model offers an appropriate trade-off. Since this is proven as an NP-hard problem, A PSO-based metaheuristic algorithm is proposed to provide near-optimal solutions for advertising in SIoT, called PSO-ASoT. Through simulations, we show that our PSO-ASoT algorithm outperforms state-of-the-art algorithms in several metrics, such as influence spread, advertising cost, and duplicated advertising rate. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Heuristic algorithms

来源：评论

学校读者我要写书评

暂无评论

A Lightweight Object Detection Algorithm for Remote Sensing Images

引用

engineering Letters 2025年第3期33卷 704-711页

作者： Hou, Donghao Zhang, Yujun Ren, Jia School of Computer and Software Engineering University of Science and Technology Liaoning Anshan114051 China School of Electronic and Information Engineering University of Science and Technology Liaoning Anshan114051 China

With the continuous advancement of satellite technology, remote sensing images has been increasingly applied in fields such as urban planning, environmental monitoring, and disaster response. However, remote sensing images often feature small target sizes and complex backgrounds, posing significant computational challenges for object detection tasks. To address this issue, this paper proposes a lightweight remote sensing images object detection algorithm based on YOLOv9. The proposed algorithm incorporates the SimRMB module, which effectively reduces computational complexity while improving the efficiency and accuracy of feature extraction. Through a dynamic attention mechanism, SimRMB is capable of focusing on important regions while minimizing background interference, and by integrating residual learning and skip connections, it ensures the stability of deep networks. To further enhance detection performance, the FasterRepNCSPELAN4 module is introduced, which employs PConv operations to reduce computational load and memory usage. It also utilizes dilated convolutions and DFC attention mechanisms to strengthen feature extraction, thereby increasing the efficiency and accuracy of object detection. Additionally, this study integrates the GhostModuleV2 module, which generates core feature maps and employs lightweight operations to create redundant features, greatly reducing the computational complexity of *** results show that on the SIMD dataset, the improved YOLOv9 model has a parameter size of 167.88 MB and GFLOPs of 208.6. Compared to the baseline YOLOv9 model (parameter size: 194.57 MB, GFLOPs: 239.0), the parameter size is reduced by 13.71%, GFLOPs are reduced by 12.72%, and detection accuracy is improved by 1.4%. These results demonstrate that the proposed lightweight YOLOv9 model effectively reduces computational overhead while maintaining excellent detection performance, providing an efficient solution for object detection tasks in resou

关键词： Satellite imagery

来源：评论

学校读者我要写书评

暂无评论

Steel Surface Defect Detection Based on YOLOv10

引用

engineering Letters 2025年第5期33卷 1220-1231页

作者： Sun, Luyu Zhang, Yujun School of Computer and Software Engineering University of Science and Technology Liaoning Anshan114051 China

To improve the accuracy of steel surface defect detection, this study proposes an improved multi-directional optimization model based on the YOLOv10n algorithm. First, we introduce innovations to the convolution (C2F) module in YOLOv10n with MSMHSA (Multi-Scale Multi-Head Self-Attention) and EMA (Enhanced Multi-Scale Attention) modules. These modules enhance learning and expression capabilities by reshaping across different scales, multiple attention heads, and channel dimensions. Second, a Multi-Dilated Channel Refinement (MDCR) module is employed. The MDCR module captures spatial features across various receptive field ranges through multiple depthwise separable convolutions, enabling more effective multi-level feature integration. Finally, a Context Aggregation module is embedded within the neck network as a generic building block for multi-head context aggregation, leveraging the inductive bias of local convolution operations to facilitate rapid convergence. Experimental results show that the improved model achieves a mean Average Precision (mAP) of 78.7% in steel surface defect detection tasks, marking a significant improvement of 4.3% over the original YOLOv10n model. In practical applications, the improved model can quickly and accurately locate and classify various steel surface defects, meeting the needs of steel surface defect detection in industrial production. © 2025, International Association of Engineers. All rights reserved.

关键词： Surface defects

来源：评论

学校读者我要写书评

暂无评论

Chinese Medical Named Entity Recognition Using Pre-Trained Language Models and an Efficient Global Pointer Mechanism

Informatica (Slovenia)

引用

Informatica (Slovenia) 2025年第19期49卷 97-114页

作者： Zhang, Xu Li, Feihong Li, Chenlong Yu, Xufeng College of Software Engineering Xiamen University of Technology Xiamen361000 China College of Computer and Information Engineering Xiamen University of Technology Xiamen361000 China

Medical Named Entity Recognition (MNER) is a critical task in medical text mining, serving as a foundation for intelligent diagnosis, disease prediction, and related tasks. However, Chinese medical texts present unique challenges, such as diverse entity classifications, varying entity lengths, and complex entity nesting, which hinder NER performance. To address these issues, we propose a Chinese MNER method leveraging pre-trained models and the efficient global pointer (EGP). Our approach incorporates three key innovations: (1) data augmentation techniques, including easy data augmentation and remote supervision with a medical entity dictionary, to address imbalanced entity types distribution and varying entity lengths;(2) a character-word fusion strategy that integrates RoBERTa-generated character vectors and Word2Vec-generated word vectors, enhancing semantic representation;and (3) an improved decoding layer based on EGP, enabling efficient recognition of both nested and flat entities while reducing computational costs. Experimentals show that our method achieves F1 scores of 75.87% and 92.77% on the CMeEE-V2 and CCKS2020 datasets, respectively, outperforming the RoBERTa-BiLSTM-CRF baseline by 3.06% and 4.38%, respectively. © 2025 Slovene Society Informatika. All rights reserved.

关键词： data augmentation EGP MNER RoBERTa Word2Vec

来源：评论

学校读者我要写书评

暂无评论

Infrared Small Target Detection with UAV Based on Convolutional Neural Networks

引用

engineering Letters 2025年第5期33卷 1505-1512页

作者： Zhang, Jingshi Zhang, Yujun School of Computer and Software Engineering University of Science and Technology Liaoning Anshan114051 China

Infrared target detection is now applied in many fields, such as medical imaging, military detection, autonomous driving, and environmental monitoring with drones. Due to the small size of these targets, complex environments, and limited contextual information, target detection is challenging. Therefore, this paper proposes an improved method for detecting small infrared targets based on the YOLOv9-c ***, we propose dynamic snake convolution (DSConv), a mechanism designed to precisely model tubular structures by adaptively concentrating on elongated and intricate curvilinear features within local regions. Second, we employ a new spatial and channel synergistic attention module (SCSA), integrating the SCSA module into the neck network of YOLOv9. The SCSA framework integrates two synergistic components, the Shared Multi-Semantic Spatial Attention (SMSA) module and the Progressive Channel Self-Attention (PCSA) module. This enhances the synergy between spatial and channel attention across multiple semantic levels. Finally, we introduce the C2F module into multi-scale dilated attention, altering its original dilation rates. C2F MSDA better aggregates semantic information, effectively reducing the redundancy of self-attention mechanisms. This improvement has shown promising results on the public HIT-UAV dataset, raising the average precision to 0.902 and providing potential technological support for the detection of small infrared targets. © 2025, International Association of Engineers. All rights reserved.

关键词： Target drones

来源：评论

学校读者我要写书评

暂无评论

Diabetic Retinopathy Image Segmentation Method Based on Fusion DenseNet and U-Net Network

引用

engineering Letters 2025年第2期33卷 418-428页

作者： Zhang, Yaxuan Xu, Yang Zhai, Jinlong University of Science and Technology LiaoNing Anshan114051 China School of Computer Science and Software Engineering University of Science and Technology LiaoNing Anshan114051 China

Diabetic Retinopathy (DR) is a common and significant complication in patients with diabetes, and severely affecting their quality of life. Image segmentation plays a crucial role in the early diagnosis and treatment of DR. However, traditional methods are limited in terms of segmentation accuracy and generalization capability. This paper proposes a novel image segmentation method for diabetic retinopathy based on the integration of DenseNet and U-Net networks. Firstly, DenseNet is utilized to replace the encoder part of U-Net. The dense connection mechanism enhances the efficiency of feature propagation and improves the feature extraction capability of the encoder. Secondly, we introduce Omni-Dimensional Dynamic Convolution (ODConv) to replace traditional convolutions. ODConv is used to handle diverse input features effectively. The model's adaptability and segmentation accuracy for various samples is improved. Finally, we integrate the Convolutional Block Attention Module (CBAM) into the decoder. By leveraging channel and spatial attention, the ability to capture key features is enhanced, and the segmentation accuracy and boundary recognition capabilities are improved. To validate the effectiveness and feasibility of the model. This paper conducts experiments on the DDR dataset and the IDRID dataset. The experimental results show that the proposed method improved the mDice score by 0.53% compared to traditional methods. It also increased the mIoU score by 1.12% over traditional methods. This indicates that the proposed method has better performance in segmenting diabetic retinopathy images. It also demonstrates better generalization ability. © 2025, International Association of Engineers. All rights reserved.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Finite-Time Sliding Mode Control of Vehicle Formation with Specified Performance

引用

engineering Letters 2025年第3期33卷 620-629页

作者： Guo, Dongyang Zhang, Zhao Zhou, Hongyan Chen, Xue-Bo School of Electronic and Information Engineering University of Science and Technology Liaoning Anshan114051 China School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan114051 China

The current urban intelligent transportation is in a rapid development stage, and coherence control of vehicle formations has important implications in urban intelligent transportation research. This article focuses on the problem of urban intelligent transportation planning, and proposes a new control method for vehicle formation regulation tracking performance control with bounded disturbances and model uncertainty, combining the multiple advantages of sliding mode control and finite time control and utilizing the universal approximation of neural network. This method can ensure that the vehicle formation can achieve the specified tracking performance and be stable in finite time instead of asymptotically stable. First, the third-order dynamics model and spacing strategy of the vehicle are given. The tracking performance of the vehicle formation is specified. Then, the control objective of the system is transformed by using the error transformation to obtain a new system. An improved sliding mode surface is designed for the transformed system. Furthermore, the universal approximation property of the neural network is utilised to overcome the parameter uncertainty in the system. The reconstruction error of neural network is handled by a robust term. In addition, the jittery vibration phenomenon of sliding mode control has been overcome. Finally, the finite-time stability of the system is analysed by constructing the Lyapunov function, simulations were performed to validate and compare with conventional control methods, simulation results show that the novel control method proposed in this paper is significantly faster than the conventional control method, validated the effectiveness of the method proposed in this paper. © 2025, International Association of Engineers. All rights reserved.

关键词： Lyapunov functions

来源：评论

学校读者我要写书评

暂无评论

Clustering and Artificial Intelligence-based Prediction of Ecologically Sustainable Species Introductions

IAENG International Journal of Computer Science

引用

IAENG International Journal of computer Science 2025年第4期52卷 1159-1168页

作者： Liu, Shuqiao Zhang, Zhao Zhou, Hongyan Chen, Xue-Bo School of Electronic and Information Engineering University of Science and Technology Liaoning Anshan114051 China School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan114051 China

There is a growing interest in sustainable ecosystem development, which includes methods such as scientific modeling, environmental assessment, and development forecasting and planning. However, due to insufficient survey data in many current development areas, development progress is delayed and stagnant. To address this situation, this paper proposes a SWOT-TOPSIS-K-Means (STK) data analysis and evaluation model to analyze ecological factors, which can realize a comprehensive and complete data analysis with fewer samples. Decision tree (DT), random forest (RF), and multilayer perceptron (MLP) neural network models were constructed from the results of this analysis, and statistical tests such as r-squared, mean absolute error, and cross-validation are used to further confirm the performance efficiency of the computational prediction models to provide real-time prediction research solutions. For this purpose, data from research scholars on species introduction in ecosystem development were selected for testing. The results show that the proposed assessment model and modeling results satisfy all accuracy-related acceptance requirements. Among them, MLP is better than DT and RF. In summary, the STK assessment model and the MLP prediction model can provide a basis for the selection and development of ecological factors. © (2025), (International Association of Engineers). All rights reserved.

关键词： Prediction models

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：