检索结果-内蒙古大学图书馆

A novel CatractNetDetect deep learning model for effective cataract classification through data fusion of fundus images

引用

Discover Artificial Intelligence 2024年第1期4卷 54页

作者： Ismail, Walaa N. Alsalamah, Hessah A. Department of Management Information Systems College of Business Al Yamamah University Riyadh11512 Saudi Arabia College of Computer and Information Sciences King Saud University Riyadh4545 Saudi Arabia Computer Engineering Department College of Engineering and Architecture Al Yamamah University Riyadh11512 Saudi Arabia

Cataracts are common eye disorders characterized by the clouding of the lens, preventing light from passing through and impairing vision. Various factors, including changes in the lens’s hydration or alterations in its proteins, may contribute to their development. Regular eye examinations conducted by an ophthalmologist or optometrist are imperative for detecting cataracts and other ocular conditions early on. Manual checks by caregivers pose several problems, including subjectivity, human error, and a lack of expertise. Biomedical fusion involves combining or linking various characteristics specific to certain diseases from different medical imaging resources. The primary objectives of this approach in disease classification are to reduce the error rate and increase the number of retrieved features. The aim of this study is to evaluate the outcomes associated with fusing visual features related to left and right eye cataract characteristics. Additionally, we investigate the impact of limited variability in deep learning models, specifically in the classification of cataract fundus versus normal fundus images. To address this issue, this study introduces CataractNetDetect, an innovative multi-label deep learning classification system that fuses feature representations from pairs of fundus images (e.g., left and right eyes) for the automatic diagnosis of various ocular disorders. Our focus is on achieving improved performance by stacking discriminative deep feature representations to combine two fundus images into a unified feature representation. Several deep learning architectures are utilized as feature descriptors, including ResNet-50, DenseNet-121, and Inception-V3, enhancing the resilience and quality of representations. Fine-tuning of these DL architectures is conducted using the ImageNet dataset, followed by an integrated stacking approach combining ResNet-50, DenseNet-121, and Inception-V3 models. The model is trained on the publicly available ODIR-5k datas

关键词： Medical imaging

来源：评论

学校读者我要写书评

暂无评论

Hyperspectral Image Classification Method Based on Global Space-Spectral Attention Mechanism

引用

Journal of Shanghai Jiaotong University (Science) 2025年 1-11页

作者： Qin, Rui Wu, Benze Liu, Xinfu Wu, Yirui College of Computer Science and Software Engineering Hohai University Nanjing211100 China Nangjing Research Institute of Electronic Engineering Nanjing210007 China College of Artificial Intelligence and Automation Hohai University Jiangsu Changzhou213200 China

In hyperspectral remote sensing imagery, pixel interactions within defined spatial extents result in the mixing of adjacent pixels. Additionally, the high similarity of adjacent spectra leads to information redundancy, which hinders the extraction of global spatial and spectral correlations. In order to solve the problems of mixed adjacent pixels and redundant adjacent spectra, this work offers a hyperspectral image classification approach that uses a global space-spectral attention mechanism. First, the proposed method’s global spatial attention module uses multi-scale dilated convolution to produce a bigger receptive field to be capable of capturing global spatial correlation and obtain unmixed pixel information. Then, the global spectral attention module designs a spectral domain partition algorithm, using the combination of regional density as well as information entropy as the threshold to divide spectrum into dispersed subsets and eliminate redundant information. The global context information for entire spectral band is fully exploited, and correlation of the global spectral information is extracted. Finally, the two modules combine to provide a global correlation of space and spectrum. Experiments demonstrate that the suggested method obtains overall accuracies of 97.28%, 94.73%, and 95.76% on the three WHU-Hi hyperspectral datasets, surpassing comparison methods. © Shanghai Jiao Tong University 2024.

关键词： Pixels

来源：评论

学校读者我要写书评

暂无评论

Contrastive Learning-based Meta-Learning Sequential Recommendation 2

Contrastive Learning-based Meta-Learning Sequential Recommen...

引用

2nd International Conference on Distributed Computing and Optimization Techniques, ICDCOT 2024

作者： Li, Naihao Wang, Jiaqi Liu, Xu Wang, Lanfeng Zhang, Long College of Computer and Information Engineering Tianjin Normal University Tianjin China College of Computer and Information Engineering Guangzhou Huali Science and Technology Vocational College Guangdong Guangzhou China

ISBN: (纸本)9798350382952

Contrastive Learning (CL) has significant practical value in the current field of recommendation systems. Contrastive learning involves learning representations by comparing and contrasting the similarities and differences between different data points or samples. Currently, contrastive learning (CL) is emerging as a promising alternative method to tackle challenges of data with low density and high levels of noise in recommendation systems. However, existing CL methods often focus on finding different augmentation operations for different datasets, which leads to a lack of generalization in the models. Additionally, these CL methods may exhibit abnormal behavior when dealing with limited resources due to the uncertainty introduced by data augmentation. Moreover, in recommendation systems, there is a well-known phenomenon where a small number of popular items account for a large portion of user interactions, while numerous cold-start items have extremely few interaction data points. Therefore, this paper proposes a more versatile contrastive learning model that mitigates the issues of data sparsity and noise, termed the Contrastive Learning with Meta-learning Framework (CLMLRec). Building upon the existing model MCLRec, this paper introduces a pre-training warm-up framework (PRWF) that includes Meta Scaling and Shift Networks, which create adjustment and displacement individually for each item. Through this framework, the paper aims to learn to pre-activate cold item embeddings and demonstrates the effectiveness of CLMLRec on commonly used datasets. © 2024 IEEE.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Cross-modal retrieval of large-scale images in social media based on spatial distribution entropy

引用

International Journal of Web Based Communities 2024年第1-2期20卷 88-101页

作者： Ding, Jie Zhao, Guotao Xu, Fang College of Technology Hubei Engineering University Xiaogan China School of Foreign Languages Hubei Engineering University Xiaogan China School of Computer and Information Science Hubei Engineering University Xiaogan China

In order to improve the cross-modal retrieval accuracy of large-scale social media images, a cross-modal retrieval method for large-scale social media images based on spatial distribution entropy is proposed. First, extract the information features of the colour and texture of the image. Then, use the image cross-modal retrieval method based on the spatial distribution entropy to calculate the spatial distribution entropy of the colour information and texture information features in the image. Finally, use the Euclidean distance to judge the space between social media images. The matching degree of the distribution entropy, according to the matching degree, is used to judge whether the image cross-modal retrieval is successful or not. The experimental results verify that the proposed method can implement comprehensive retrieval according to the specific characteristics of the retrieved images, and the matching degree of image retrieval is greater than 95%, and the retrieval accuracy is high. Copyright © 2024 Inderscience Enterprises Ltd.

关键词： Image retrieval

来源：评论

学校读者我要写书评

暂无评论

Cognitive agent based fault tolerance in ubiquitous networks: a machine learning approach

引用

International Journal of information Technology (Singapore) 2024年第4期16卷 2363-2377页

作者： Bhajantri, Lokesh B. Ayyannavar, Vasudha V. Department of Information Science and Engineering Basaveshwar Engineering College Karnataka Bagalkote India Department of Computer Science and Engineering Basaveshwar Engineering College Bagalkote VTU Karnataka Belagavi India

Ubiquitous Networks play an essential role in accessing ubiquitous computing services at anytime, anywhere, and anyplace through computing nodes of heterogeneous networks. Nowadays, ubiquitous network faces various issues related to fault management or tolerance in a real world environment. In this type of network, fault tolerance is one of the vital issues due to dynamic and uncertain character of ubiquitous networks with various constraints of nodes and networks to maximize network lifetime, availability, and reliability of the networks. In this paper, it is proposed that a cognitive agent based fault tolerance system using reinforcement learning algorithm to provide efficient ubiquitous services to the users over the networks. The main objective of the proposed work is to emphasize the analysis, identification, and recovery of various faults under resource constraints of ubiquitous networks in terms of node, memory, load, and network level faults (link level fault and packet error) using a cognitive agent based machine learning algorithm. Also, the proposed work performs better fault tolerance than conventional methods (without using cognitive agents) by considering fault detection rate, fault recovery rate, fault recovery time, packet delivery ratio, agent computation overhead, and energy consumption for performing the desired set of tasks and services ubiquitously in the ubiquitous computing environment. © The Author(s), under exclusive licence to Bharati Vidyapeeth's Institute of computer Applications and Management 2024.

关键词： Cognitive agents Heterogeneous network systems Reinforcement learning Ubiquitous networks Ubiquitous service systems

来源：评论

学校读者我要写书评

暂无评论

Advanced Bearing Fault Diagnosis Using Cuckoo Optimization and KAN Algorithms 4

Advanced Bearing Fault Diagnosis Using Cuckoo Optimization a...

引用

4th International Conference on Electronic information engineering and computer Science, EIECS 2024

作者： Wang, Lingchao Ai, Qiang Yan, Hailong Hao, Ming Li, Xiaoxin School of Information Engineering Liaoning Institute of Science and Engineering Jinzhou China College of Computer Qinghai Normal University Xining China Software College Liaoning Technical University Huludao China

ISBN: (纸本)9798331531409

This study proposes a bearing fault diagnosis method that combines the Cuckoo Optimization Algorithm (COA) with the KAN algorithm. COA, as an intelligent optimization algorithm, is primarily used to find the optimal hyperparameters of the model, thereby enhancing the efficiency and robustness of the diagnosis model. The KAN algorithm is employed to build an intelligent classification model, achieving accurate recognition of multiple fault modes. Experimental results show that the proposed method maintains high diagnostic accuracy even with limited data, providing an effective solution for bearing fault diagnosis. © 2024 IEEE.

关键词： Fault detection

来源：评论

学校读者我要写书评

暂无评论

Advancing Deepfake Detection Using Xception Architecture:A Robust Approach for Safeguarding against Fabricated News on Social Media

引用

computers, Materials & Continua 2024年第12期81卷 4285-4305页

作者： Dunya Ahmed Alkurdi Mesut Cevik Abdurrahim Akgundogdu College of Information Engineering Al-Nahrain UniversityBaghdad10011Iraq Department of Electrical and Computer Engineering Altinbas UniversityIstanbul34000Turkey Electrical and Electronics Engineering Department Istanbul University-CerrahpasaIstanbul34320Turkey

Deepfake has emerged as an obstinate challenge in a world dominated by ***,the authors introduce a new deepfake detection method based on Xception *** model is tested exhaustively with millions of frames and diverse video clips;accuracy levels as high as 99.65%are *** are the main reasons for such high efficacy:superior feature extraction capabilities and stable training mechanisms,such as early stopping,characterizing the Xception *** methodology applied is also more advanced when it comes to data preprocessing steps,making use of state-of-the-art techniques applied to ensure constant *** an ever-rising threat from fake media,this piece of research puts great emphasis on stringent memory testing to keep at bay the spread of manipulated *** also justifies better explanation methods to justify the reasoning done by the model for those decisions that build more trust and *** ensemble models being more accurate have been studied and examined for establishing a possibility of combining various detection frameworks that could together produce superior ***,the study underlines the need for real-time detection tools that can be effective on different social media sites and digital ***,protecting privacy,and public awareness in the fight against the proliferation of deepfakes are important *** significantly contributing to the advancements made in the technology that has actually advanced detection,it strengthens the safety and integrity of the cyber world with a robust defense against ever-evolving deepfake threats in ***,the findings generally go a long way to prove themselves as the crucial step forward to ensuring information authenticity and the trustworthiness of society in this digital world.

关键词： Deepfake Detection Xception architecture data processing image processing intelligent information systems social media security

来源：评论

学校读者我要写书评

暂无评论

Smart automated highway lighting system using IoT: a survey

引用

Energy Informatics 2024年第1期7卷 76页

作者： Achar, Tejaswini Eshwar Rekha, C. Shreyas, J. Department of Computer Science and Engineering Government Engineering College Koppal Talakal India Department of Information Science and Engineering Bapuji Institute of Engineering and Technology Davanagere India Department of Information Technology Manipal Institute of Technology Bengaluru Manipal Academy of Higher Education Manipal India

Efficient highway lighting is crucial for ensuring road safety and reducing energy consumption and costs. Traditional highway lighting systems rely on timers or simple photosensors, leading to inefficient operation by illuminating lights when not needed or failing to adjust to changing conditions. The emergence of the Internet of Things (IoT) and related technologies has enabled the development of smart automated highway lighting systems that can dynamically control illumination levels based on real-time data. This paper provides a comprehensive review of the current state-of-the-art in smart automated highway lighting systems employing IoT technologies. Key components, communication protocols, data processing techniques, and lighting control strategies are discussed. The integration of renewable energy sources and energy storage systems is explored for environmentally sustainable operations. Practical implementation case studies are analyzed to highlight benefits and challenges. Open research issues and future directions for further enhancements are identified. © The Author(s) 2024.

关键词： Street lighting

来源：评论

学校读者我要写书评

暂无评论

Experiment Teaching Case Design of computer I/O Channel Virtual Simulation 18th

Experiment Teaching Case Design of Computer I/O Channel Vir...

引用

18th International Conference on computer Science and Education, ICCSE 2023

作者： Zhang, Dongdong Cheng, Xinyuan Gao, Zhen Qin, Guofeng Huang, Jie Wang, Lisheng College of Electronics and Information Engineering Tongji University Shanghai China National Computer and Information Technology Experiment Teaching Demonstration Center Tongji University Shanghai China College of Software Engineering Tongji University Shanghai China

ISBN: (纸本)9789819707904

The principle of computer composition is an important basic course for undergraduates majoring in computer science, in which the I/O channel control method in the input/output system is mainly applied to mainframes. However, mainframes that support I/O channel design experiments are expensive, complex and costly to operate and maintain, which makes it impossible to carry out hands-on experiments for large-scale students. In this paper, to break the situation that this part of content is mainly taught through textbook in traditional teaching, we design a virtual simulation experiment of computer I/O channel, which is an effective supplement to the practical teaching of computer composition principle. The experimental case has been applied in computer science and technology, information security and software engineering in Tongji University and other colleges and universities, and has been awarded Chinese national first-class course (Virtual Simulation Class). The experimental system has been opened and shared on Chinese national virtual simulation experiment teaching project sharing service platform (***). The experiment website is http://***/details/2020?id=7256. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： computer software

来源：评论

学校读者我要写书评

暂无评论

Speech Emotion Recognition based on Multi COATTENTION Acoustic Feature Fusion 2

Speech Emotion Recognition based on Multi COATTENTION Acoust...

引用

2nd International Conference on Data Science and information System, ICDSIS 2024

作者： Song, Hongchen Kang, Kuo Wei, Yuhua Zhang, Hengyuan Zhang, Long Tianjin Normal University College Of Computer And Information Engineering Tianjin China Guangzhou Huali Science And Technology Vocational College College Of Computer Information Engineering Guangdong Guangzhou China

ISBN: (纸本)9798350381665

Speech Emotion Recognition (SER) is a technology aimed at extracting emotional information from speech signals, thereby discerning the emotional state of the speaker. Its objective is to equip machines with empathetic capabilities akin to those of humans. The ability to extract high-quality, deep, and comprehensive speech representations from audio often determines the robustness of emotion analysis. In this paper, we focus on modeling speech using spectrograms, MFCCs (Mel-frequency cepstral coefficients), and raw waveforms. We employ CNNs (Convolutional Neural Networks), BiGRUs (Bidirectional Gated Recurrent Units), and self-supervised Wav2Vec-based models [1] to extract deep acoustic features, each applied separately. Taking into account the complementarity among different acoustic features, we introduce a Multi CO-ATTENTION Mechanism to enhance the acoustic characteristics. Additionally, feature fusion is performed at the final stage of the model. We evaluated our proposed method on the IEMOCAP dataset and conducted ablation studies, achieving promising results. © 2024 IEEE.

关键词： Quality control

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：