检索结果-内蒙古大学图书馆

Unveiling anomalies: transformative insights from transformer-based autoencoder models

International Journal of Computers and Applications 2025年第1期47卷 29-44页

作者： Dehkordi, Sajad Balali Nasri, Saeed Dami, Sina Department of Electrical Engineering Najaf Abad Branch Islamic Azad University Najaf Abad Iran Digital Processing and Machine Vision Research Center Najaf Abad Branch Islamic Azad University Najaf Abad Iran Department of Computer Engineering West Tehran Branch Islamic Azad University Tehran Iran

Nowadays, Intrusion Detection Systems (IDS) play a critical role in safeguarding networks by identifying and mitigating unauthorized access and malicious activities, yet face challenges in accurately detecting sophisticated threats while minimizing false positives. In this paper, we present a comprehensive analysis of a Transformer-based Autoencoder model for network intrusion detection using the NSL-KDD dataset. We evaluate the model's anomaly detection performance based on key metrics such as Mean Squared Error (MSE), Precision-Recall Area Under Curve (AUC), and Root Mean Squared Error (RMSE) on the test set. Our findings reveal promising results, with the model achieving low MSE, high AUC, and relatively low RMSE values, indicating its effectiveness in detecting anomalies. Additionally, we provide an in-depth examination of the model architecture, highlighting the role of encoder and decoder layers, dropout regularization, and optimization techniques such as Adam optimizer. Our analysis sheds light on the efficacy of Transformer-based Autoencoder models for network intrusion detection tasks, offering insights into their architectural design and performance evaluation. The novelty of this study lies in the implementation of a Transformer-based Autoencoder architecture for network intrusion detection, showcasing its superior performance in anomaly detection compared to traditional methods. © 2024 Informa UK Limited, trading as Taylor & Francis Group.

关键词： Network intrusion

来源：评论

学校读者我要写书评

暂无评论

MSRFormer: Hybrid Scale Self-Attention and Local Fast Convolution Transformer for Facial Expression Recognition

MSRFormer: Hybrid Scale Self-Attention and Local Fast Convol...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal processing, ICASSP 2025

作者： Shen, Zi-Qiang Tang, Yu-Yi Yan, Jun-Feng Li, Yang Zhao, Guo-Ying Hunan University of Chinese Medicine China Hunan Digital Chinese Medicine Engineering Technology Research Center China Center for Machine Vision Research Department of Computer Science and Engineering University of Oulu Finland

ISBN: (纸本)9798350368741

With the emergence of the Transformer architecture, the accuracy of deep learning within the domain of facial emotion recognition has seen further enhancement. However, Transformer comes with increased training complexity and time due to the large parameter count. Additionally, the global receptive field in Transformer's attention leads to unnecessary computations for features with limited spatial extent in image sentiment analysis. In this paper we presents a MSRFormer model, which combines Hybrid-scale self-attention and local fast convolution to address existing issues. The Hybrid-scale self-attention enables precise focus on salient regions of the image, where key features are located. A Local Fast Convolution module was integrated into the MLP head of the Transformer model, enhancing training speed and reducing parameters while maintaining feature learning. we conducted experiments on FERPlus, RAF-DB, and KDEF datasets and confirming its strong performance and broad applicability in facial expression recognition. © 2025 IEEE.

关键词： facial expression recognition hybrid scale self attention local fast convolution transformer

来源：评论

学校读者我要写书评

暂无评论

Chromosome analysis using a hybrid deep CNN and structural feature-based grouping model

引用

Multimedia Tools and Applications 2025年 1-30页

作者： Isfahani, Farahnaz Peiravi Pourghassem, Hossein Mahdavi-Nasab, Homayoun Naghsh, Alireza Department of Electrical Engineering Najafabad Branch Islamic Azad University Najafabad Iran Digital Processing and Machine Vision Research Center Najafabad Branch Islamic Azad University Najafabad Iran

Chromosome analysis and classification are essential in clinical applications to diagnose various structural and numerical abnormalities. Recently, karyotype analysis using intelligent image processing methods, especially deep learning, has attracted significant attention as a genetic abnormality test. This paper presents a novel chromosome classification algorithm that uses high-level features extracted from deep convolutional neural networks (DCNN) along with morphological features designed to identify and modify the classes of misclassified chromosomes. Initially, chromosomes are classified using a DCNN. Some structural features, such as centromere and banding profile, are then extracted to group chromosomes again. Based on the results of the two preceding methods, a decision strategy is utilized to identify misclassified chromosomes. Here, a final DCNN-based strategy is introduced to assign misclassified chromosomes to the associated classes. The proposed method can be used in parallel with other chromosome classification methods to modify misclassified chromosomes and promote the accuracy of the classification. Evaluation results show that the proposed algorithm outperforms relevant state-of-the-art algorithms regarding the classification precision and accuracy of 99.66 and 96.52%, respectively. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Mathematical Formulas-Based Scientific Literature Retrieval with Heterogeneous Network Semantic Enhancement 25th

Mathematical Formulas-Based Scientific Literature Retrieval ...

引用

25th International Conference on Web Information Systems Engineering, WISE 2024

作者： He, Xin School of Cyber Security and Computer Hebei University Baoding071002 China Institute of Intelligent Image and Document Information Processing Hebei University Baoding071002 China Hebei Machine Vision Engineering Research Center Hebei University Baoding071002 China

ISBN: (纸本)9789819605781

As the critical elements of scientific literature, mathematical formulas contribute significantly to the effectiveness of literature retrieval. Existing formula-based retrieval methods of scientific literature primarily focus on the number of matching symbols between formulas and often neglect the semantic relevance which leads to suboptimal retrieval outcomes. Additionally, existing literature retrieval methods frequently ignore the ontological relationships between literature, resulting in potential semantic information loss. To address these challenges, we propose a two-stage scientific literature retrieval model named SC-HNE (Symbol Contribution-Heterogeneous Network Enhancement). First, we introduce a Symbol Contribution Aware (SCA) module that calculates the contribution of each symbol based on its hierarchical position in the Operator Tree (OPT), and the symbols with the higher contributions will be paid more attention during the process of formula matching. Second, a heterogeneous network for scientific literature is constructed which uses Heterformer to enrich literature node embeddings with network-aware information and enhance literature representation. Finally, the similarity scores from two modules are combined to rank the top-K scientific literature. Experiments on the DBLP-Citation-network dataset demonstrated the effectiveness of the proposed method through MAP and NDCG metrics. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Trees (mathematics)

来源：评论

学校读者我要写书评

暂无评论

MSRFormer: Hybrid Scale Self-Attention and Local Fast Convolution Transformer for Facial Expression Recognition

MSRFormer: Hybrid Scale Self-Attention and Local Fast Convol...

引用

International Conference on Acoustics, Speech, and Signal processing (ICASSP)

作者： Zi-Qiang Shen Yu-Yi Tang Jun-Feng Yan Yang Li Guo-Ying Zhao Hunan University of Chinese Medicine China Hunan Digital Chinese Medicine Engineering Technology Research Center China Department of Computer Science and Engineering Center for Machine Vision Research University of Oulu Finland

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

关键词： Training Representation learning Sentiment analysis Accuracy Head Convolution Face recognition Speech recognition Transformers Speech processing

来源：评论

学校读者我要写书评

暂无评论

Clinical Validation of the Saadat NIBP Module According to the BS en ISO 81060-2 Protocol 17

Clinical Validation of the Saadat NIBP Module According to t...

引用

17th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology, ECTI-CON 2020

作者： Haghayegh, Mohammadamin Behzadfar, Neda Shahgholian, Ghazanfar Najafabad Branch Islamic Azad University Digital Processing and Machine Vision Research Center Najafabad Iran

ISBN: (纸本)9781728164861

This study was conducted according to the British Standard EN ISO 81060-2 guidelines. The aim was to validate the Saadat non-invasive blood pressure (NIBP) module accuracy and reliability against manual auscultatory general practitioners (GP). readings. A total of 298 measurements and comparison procedures were performed on 95 adults without heart disease under the supervision of two GPs. In order to represent the relationships between the test device and the reference method, the Bland-Altman graphical plotting method was used. The mean differences and standard deviations (mean ± SD differences) between the readings of the Saadat NIBP module and determination of GPs as the reference method for systolic and diastolic were exceptionally close with: -2.22 ± 6.51 mmHg and -3.31 ± 6.27 mmHg respectively. The Saadat NIBP module fulfilled the BS EN ISO 81060-2 requirements, which states that the mean ± SD of lower than 5 ± 8 mmHg can be recommended for clinical use. © 2020 IEEE.

关键词： Blood pressure

来源：评论

学校读者我要写书评

暂无评论

Early detection of diabetes through transfer learning-based eye (vision) screening and improvement of machine learning model performance and advanced parameter setting algorithms

arXiv

引用

arXiv 2025年

作者： Yousefi, Mohammad Reza Bakrani, Ali Dehghani, Amin Department of Electrical Engineering Najafabad Branch Islamic Azad University Najafabad Iran Digital Processing and Machine Vision Research Center Najafabad Branch Islamic Azad University Najafabad Iran School of Electrical and Computer Engineering University of Tehran Iran

Diabetic Retinopathy (DR) is a serious and common complication of diabetes, caused by prolonged high blood sugar levels that damage the small retinal blood vessels. If left untreated, DR can progress to retinal vein occlusion and stimulate abnormal blood vessel growth, significantly increasing the risk of blindness. Traditional diabetes diagnosis methods often utilize convolutional neural networks (CNNs) to extract visual features from retinal images, followed by classification algorithms such as decision trees and k-nearest neighbors (KNN) for disease detection. However, these approaches face several challenges, including low accuracy and sensitivity, lengthy machine learning (ML) model training due to high data complexity and volume, and the use of limited datasets for testing and evaluation. This study investigates the application of transfer learning (TL) to enhance ML model performance in DR detection. Key improvements include dimensionality reduction, optimized learning rate adjustments, and advanced parameter tuning algorithms, aimed at increasing efficiency and diagnostic accuracy. The proposed model achieved an overall accuracy of 84% on the testing dataset, outperforming prior studies. The highest class-specific accuracy reached 89%, with a maximum sensitivity of 97% and an F1-score of 92%, demonstrating strong performance in identifying DR cases. These findings suggest that TL-based DR screening is a promising approach for early diagnosis, enabling timely interventions to prevent vision loss and improve patient outcomes. © 2025, CC BY.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Fixed-time multilayer neural network-based leader–follower formation control of autonomous surface vessels with limited field-of-view sensors and saturated actuators

引用

Neural Computing and Applications 2024年 1-21页

作者： Naderolasli, Amir Shojaei, Khoshnam Chatraei, Abbas Department of Electrical Engineering Najafabad Branch Islamic Azad University Najafabad Iran Digital Processing and Machine Vision Research Center Najafabad Branch Islamic Azad University Najafabad Iran

This paper proposes a fixed-time multilayer neural network-based formation control problem for the autonomous surface vessels based on the relative distance and orientation angle constraints. The proposed approach only expends the measurements of comparative distance and orientation angle sensors with a limited field-of-view (FOV). The proposed strategy is able to obtain the perfect trajectory tracking performance when the motion ability of all the vessels is restricted into a predefined region owing to the limited FOV limitations. An asymmetric time-varying barrier Lyapunov function is efficiently utilized to cope with the limited FOV constraints. A multilayer neural network is efficiently applied to estimate the model uncertainties and unmodeled dynamics through the online updating of weight matrices. The suggested controller preserves both the comparative distance and orientation angles between consecutive vessels inside the predefined constraints, and the state errors converge to small residual sets around the zero in a fixed time. This feature accelerates the convergence speed and modifies the transient performance of the trajectory tracking control for all the vessels in the formation construction. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Trajectories

来源：评论

学校读者我要写书评

暂无评论

Computer vision-based apple grading for golden delicious apples based on surface features

引用

Information processing in Agriculture 2017年第1期4卷 33-40页

作者： Payman Moallem Alireza Serajoddin Hossein Pourghassem Department of Electrical Engineering Faculty of EngineeringUniversity of IsfahanHezarjerib StreetIsfahanIran Applied Image and Signal Processing Research Group Research Center for Signal Processing and Intelligent SystemsUniversity of IsfahanHezarjerib StreetIsfahanIran Department of Electrical Engineering Najafabad BranchIslamic Azad UniversityNajafabadIsfahanIrand Digital Processing and Machine Vision Research Center Najafabad BranchIslamic Azad UniversityNajafabadIran

In this paper,a computer vision-based algorithm for golden delicious apple grading is proposed which works in six ***-apple pixels as background are firstly removed from input ***,stem end is detected by combination of morphological methods and Mahalanobis distant *** region is also detected by applying K-means clustering on the Cb component in YCbCr color *** that,defects segmentation is achieved using Multi-Layer Perceptron(MLP)neural *** the next step,stem end and calyx regions are removed from defected regions to refine and improve apple grading ***,statistical,textural and geometric features from refined defected regions are ***,for apple grading,a comparison between performance of Support Vector machine(SVM),MLP and K-Nearest Neighbor(KNN)classifiers is *** is done in two manners which in the first one,an input apple is classified into two categories of healthy and *** the second manner,the input apple is classified into three categories of first rank,second rank and rejected *** both grading steps,SVM classifier works as the best one with recognition rate of 92.5%and 89.2%for two categories(healthy and defected)and three quality categories(first rank,second rank and rejected ones),among 120 different golden delicious apple images,respectively,considering K-folding with K=***,the accuracy of the proposed segmentation algorithms including stem end detection and calyx detection are evaluated for two different apple image databases.

关键词： Golden delicious apple Grading Computer vision Segmentation Classification

来源：评论

学校读者我要写书评

暂无评论

The Change in Individual Alpha Frequency During Neurofeedback Training 30

The Change in Individual Alpha Frequency During Neurofeedbac...

引用

30th International Conference on Electrical Engineering, ICEE 2022

作者： Dorvashi, Maryam Behzadfar, Neda Islamic Azad University Najafabad Branch Department of Electrical Engineering Najafabad Iran Islamic Azad University Digital Processing and Machine Vision Research Center Najafabad Branch Najafabad Iran

ISBN: (纸本)9781665480871

The Individual Alpha Frequency (IAF) is related to cognitive functions corresponding to one's performance. This study aimed to examine the effect of Neurofeedback training (NFT) protocols on EEG signal's IAF of twelve individuals. The results showed increased IAF in all participants over the training process. Furthermore, the increase in IAF was positively correlated with the increase in Working Memory (WM) performance. This study demonstrated that one could change the Alpha band frequency range by altering the IAF using Neurofeedback. © 2022 IEEE.

关键词： Training Electrical engineering Protocols Electroencephalography Neurofeedback

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：