检索结果-内蒙古大学图书馆

Journal of Network Intelligence 2024年第1期9卷 225-238页

作者： Zhang, Jin-Quan Xu, Hui-Xin Ma, Yun-Shen Ni, Li-Na College of Computer Science and Engineering Shandong University of Science and Technology Qingdao266590 China

With the advent of the fourth industrial revolution, data ushered in explosive growth. Federated learning can protect users’ privacy and raw data from being known by third parties. Its client data is only trained locally and will not be uploaded to the server. However, the need to continuously transmit between the server and the client in federated learning increases communication overhead and communication delay, and the consumption of resources is very large. In order to improve the communication efficiency of the federation, we proposed an FedQT federated learning model compression algorithm based on pruning and quantization. Firstly, the fast sorting is used to preprocess the trained parameters. Secondly, the middle part of the processed parameters is averaged to obtain the pruning threshold avg(ω). Then, according to the obtained new threshold standard, the model performs Top Avg pruning based on median comparison. Finally, the k − means algorithm is used to cluster the parameters for further compression. In the simulation experiment, FedQT is compared with uncompressed and other compression algorithms, and the time complexity is analyzed. It is proved that FedQT can compress the trained model parameters to improve its communication efficiency while reducing the time overhead. The model can be compressed by more than one thousand times, and the model accuracy can be maintained above 97%, which is enough to prove that FedQT algorithm has good performance in the compression framework of federated learning. © 2024.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

CMDCF: an effective cross-modal dense cooperative fusion network for RGB-D SOD

引用

Neural Computing and Applications 2024年第23期36卷 14361-14378页

作者： Jia, XingZhao Zhao, WenXiu Wang, YuMei DongYe, ChangLei Peng, YanJun College of Computer Science and Engineering Shandong University of Science and Technology Qingdao266590 China

The success of vision transformer demonstrates that the transformer structure is also suitable for various vision tasks, including high-level classification tasks and low-level dense prediction tasks. Salient object detection (SOD) is a pixel-level dense prediction task that simulates the most salient objects in human visual recognition scenarios. In recent years, depth images have been widely used for salient object detection. Compared with RGB SOD, the key point of RGB-D SOD is the effective fusion of depth information. As RGB-D SOD requires extracting depth features and fusing cross-modal information, additional computation is involved. However, except for lightweight models, most RGB-D SOD methods tend to obtain better prediction maps by consuming more computational resources. We propose a cross-modal dense cooperative fusion net, which provides state-of-the-art performance with less computation and parameters. We take advantage of the ability of the transformer structure to model long sequence dependencies to extract saliency features from RGB images. Since there is less information in the depth image than in the RGB image, it is not necessary to use the same structure in the depth stream. For the sake of reducing parameters and computation, we consider the asymmetric architecture. It is enough to meet our needs that deep features extracted by lightweight MobileV2Net. Our decoder can perform dense cooperative fusion of cross-modal information while decoding features. It can both effectively fuse cross-modal information and save computation. Comprehensive experiments on multiple benchmark datasets for RGB-D SOD show that compared with SOTA methods, our method performs much better with less computation and parameters. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

TDCL: Dense Semantic Contrastive Learning for Vision-Language Tracking 27

TDCL: Dense Semantic Contrastive Learning for Vision-Languag...

引用

27th European Conference on Artificial Intelligence, ECAI 2024

作者： Wang, Zheng He, Xiankang Lan, Kaiyang Cui, Ying Guo, Dongyan College of Computer Science and Technology College of Software Zhejiang University of Technology China

ISBN: (纸本)9781643685489

Traditional single-object tracking tasks are undergoing a new wave of transformation, especially with the emergence of the lack of semantics, which has led to the rise of the vision-language tracking task. However, previous approaches that combine the visual tracker with natural language descriptions tend to rely on a global representation of the text description, considering less about the fine-grained connections between the text description and the visual appearance. This paper proposes to utilize a bi-directional cross-attention module to capture the connections between language and visual features, which are further projected as dense semantic representations for alignment. In order to keep the semantic consistency between the search region and the coupled natural language and align the fused feature, this paper proposes a novel dense semantic contrastive learning loss to bridge the semantic gap between text and visual modalities and align them in a dense form. The proposed framework achieves promising results in tracking datasets that contain natural language descriptions, such as TNL2K, and OTB99-LANG. Our approach provides a novel solution for representing and aligning cross-modal information for the single object tracking task and may inspire further research in this field. © 2024 The Authors.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Redundant Transmission Control Algorithm for Information-Centric Vehicular IoT Networks

引用

computers, Materials & Continua 2023年第8期76卷 2217-2234页

作者： Abdur Rashid Sangi Satish Anamalamudi Mohammed SAlkatheiri Murali Krishna Enduri Anil Carie Mohammed AAlqarni Department of Computer Science College of Science and TechnologyWenzhou-Kean UniversityWenzhou325060China Department of Computer Science Engineering SRM University APAmaravati522502India Department of Cybersecurity College of Computer Science&EngineeringUniversity of JeddahJeddahSaudi Arabia Department of Software Engineering College of Computer Science&EngineeringUniversity of JeddahJeddahSaudi Arabia

Vehicular Adhoc Networks(VANETs)enable vehicles to act as mobile nodes that can fetch,share,and disseminate information about vehicle safety,emergency events,warning messages,and passenger ***,the continuous dissemination of information fromvehicles and their one-hop neighbor nodes,Road Side Units(RSUs),and VANET infrastructures can lead to performance degradation of VANETs in the existing hostcentric IP-based ***,Information Centric Networks(ICN)are being explored as an alternative architecture for vehicular communication to achieve robust content distribution in highly mobile,dynamic,and errorprone *** ICN-based Vehicular-IoT networks,consumer mobility is implicitly supported,but producer mobility may result in redundant data transmission and caching inefficiency at intermediate vehicular *** paper proposes an efficient redundant transmission control algorithm based on network coding to reduce data redundancy and accelerate the efficiency of information *** proposed protocol,called Network Cording Multiple Solutions Scheduling(NCMSS),is receiver-driven collaborative scheduling between requesters and information sources that uses a global parameter expectation deadline to effectively manage the transmission of encoded data packets and control the selection of information *** results for the proposed NCMSS protocol is demonstrated to analyze the performance of ICN-vehicular-IoT networks in terms of caching,data retrieval delay,and end-to-end application *** end-to-end throughput in proposed NCMSS is 22%higher(for 1024 byte data)than existing solutions whereas delay in NCMSS is reduced by 5%in comparison with existing solutions.

关键词： Caching data dissemination redundancy control ICN-vehicular IoT networks

来源：评论

学校读者我要写书评

暂无评论

Enhancing Predictive Capabilities for Cyber Physical Systems Through Supervised Learning

Informatica (Slovenia)

引用

Informatica (Slovenia) 2025年第16期49卷 77-86页

作者： Dhanalakshmi, B. Tamije Selvy, P. Department of Computer Science and Engineering Dr.N.G. P Institute of technology India Department of Computer Science and Engineering Hindusthan College of Engineering and Technology India

The rapid advancement and proliferation of Cyber-Physical Systems (CPS) have led to an exponential increase in the volume of data generated continuously. Efficient classification of this streaming data is crucial for predicting system behaviors and enabling proactive decision-making. This research aims to extract actionable knowledge from the continuous data streams of CPS and predict their behavior using advanced supervised learning algorithms. The predictions facilitate timely interventions and necessary actions within the interconnected physical network. The background of this work lies in the intersection of CPS, machine learning, and data stream mining. Traditional batch processing methods are inadequate for real-time analysis of CPS data due to their inherent latency and computational inefficiency. This research employs state-of-the-art techniques for real-time data processing, including incremental learning, sliding window models, and ensemble methods tailored for streaming data. Our approach differs from existing works by focusing on a comprehensive framework that integrates real-time data ingestion, preprocessing, feature extraction, and model updating in a seamless pipeline. Unlike previous studies that often rely on static datasets and offline analysis, our method ensures continuous learning and adaptation to evolving data patterns. Comparative analysis with existing techniques demonstrates superior performance in terms of accuracy, latency, and scalability. Specifically, our models achieved an average classification accuracy of 92%, with a precision of 90%, recall of 89%, and an F1 score of 89.5%. These metrics indicate significant improvements over traditional batch processing methods, which typically lag in responsiveness and adaptability. This research provides a robust and efficient solution for the real-time classification of streaming data from CPS, enhancing the system's ability to predict behaviors and take necessary actions promptly. © 2025 Slov

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Blockchain Enabled Metaheuristic Cluster Based Routing Model for Wireless Networks

引用

computer Systems science & engineering 2023年第2期44卷 1233-1250页

作者： R.M.Bhavadharini S.Karthik Department of Computer Science and Engineering Easwari Engineering CollegeChennai600089India Department of Computer Science and Engineering SNS College of TechnologyCoimbatore641035India

With recent advancements made in wireless communication techniques,wireless sensors have become an essential component in both data collection as well as tracking *** Sensor Network(WSN)is an integral part of Internet of Things(IoT)and it encounters different kinds of security *** is designed as a game changer for highly secure and effective digital ***,the current research paper focuses on the design of Metaheuristic-based Clustering with Routing Protocol for Blockchain-enabled WSN abbreviated as *** proposed MCRP-BWSN technique aims at deriving a shared memory scheme using blockchain technology and determine the optimal paths to reach the destination in clustered *** MCRP-BWSN technique,Chimp Optimization Algorithm(COA)-based clustering technique is designed to elect a proper set of Cluster Heads(CHs)and organize the selected *** addition,Horse Optimization Algorithm(HOA)-based routing technique is also presented to optimally select the routes based onﬁtness ***,HOA-based routing technique utilizes blockchain technology to avail the shared mem-ory among nodes in the *** nodes are treated as coins whereas the ownership handles the sensor nodes and Base Station(BS).In order to validate the enhanced performance of the proposed MCRP-BWSN technique,a wide range of simulations was conducted and the results were examined under different *** on the performance exhibited in simulation outcomes,the pro-posed MCRP-BWSN technique has been established as a promising candidate over other existing techniques.

关键词： Wireless networks blockchain routing protocol clustering security metaheuristics

来源：评论

学校读者我要写书评

暂无评论

Semi-Supervised Multi-Task Deep Learning for WiFi Fingerprint Database Construction in Building-Scale Localization

引用

IEEE Transactions on Consumer Electronics 2024年第1期71卷 488-500页

作者： Wang, Chun Luo, Juan Yin, Luxiu Li, Chuang Huang, Wenbin Liang, Wei Li, Kuan-Ching Hunan Normal University College of Information Science and Engineering China Hunan University College of Computer Science and Electronic Engineering China Hunan University of Science and Technology School of Computer Science and Engineering China Nanjing University of Information Science and technology School of Computer Science China

WiFi-based indoor positioning has emerged as a crucial technology for enabling smart consumer electronic applications, particularly in large-scale buildings. The construction of WiFi fingerprint databases using received signal strength (RSS) is foundational due to its widespread deployment. However, achieving high positioning accuracy typically requires labor-intensive and time-consuming site surveys. While recent crowdsourcing methods have facilitated the collection of numerous RSS samples, these samples frequently lack labels and reliability in multi-scale building *** this paper, we design a novel semi-supervised and multi-task mean-teacher model (MTMT-DNN) to annotate crowdsourcing unlabeled multi-scale fingerprint samples. This method enables the construction of a comprehensive fingerprint database without requiring intensive manual effort or compromising positioning accuracy. Our key idea is to first develop a multi-task Deep Neural Network (MT-DNN) for simultaneously annotating building, floor, and intra-floor coordinate labels by leveraging their complementary information. Then we employ the mean-teacher semi-supervised learning to leverage additional unlabeled fingerprint data for further improving the annotating performance and reducing intensive manual effort. Finally, we train the MTMT-DNN model by developing two multi-task loss functions and ensuring consistency between them, thereby enhancing the reliability of the annotated crowdsourced fingerprints. We conducted real-world experiments in a 20,000 m2 site encompassing three multi-story buildings. The results demonstrate that our proposed method significantly reduces the workload of manually collecting labeled fingerprint samples. With only 20% of labeled fingerprints collected, we achieve 99% average annotation accuracy for building and floor labels and an average coordinates annotation error within 4 m. © 1975-2011 IEEE.

关键词： Floors

来源：评论

学校读者我要写书评

暂无评论

Morlet wavelet transformation based deep similarity structured neural learning for image quality assessment

引用

Multimedia Tools and Applications 2025年第16期84卷 16789-16807页

作者： Balakrishnan, N. Robinson, Y. Harold Department of MCA Sona College of Technology Salem India Department of Computer Science and Engineering Francis Xavier Engineering College Tirunelveli India

Quality assessment is a key problem to be resolved in image processing. Few research works have been designed to analyze the quality of images using different techniques. However, the accuracy involved during the process of image quality assessment was not sufficient. A novel method Morlet Wavelet Transformation Based Deep Similarity Structured Neural Learning (MWT-DSSNL) is proposed to enhance the performance of image quality assessment with minimal peak signal-to-noise ratio. The MWT-DSSNL Method is based on artificial neural networks with representation learning. The MWT-DSSNL Method initially gets the number of images from a given dataset as input. The MWT-DSSNL Method uses multiple layers to extract higher-level features from the input images. The MWT-DSSNL Method is a Feed-Forward network where each layer employs the output from the previous layer as input. The MWT-DSSNL Method is designed based on a biological neural network of the human brain. Contrary to the existing method, the MWT-DSSNL Method uses multiple hidden layers and Morlet wavelet transformation to deeply analyze input images and thereby extract features such as luminance, contrast, and structure with a minimal amount of time consumption. By considering the discovered features, finally MWT-DSSNL Method determines structural similarity and thereby exactly identifies the quality of input images with a lower time. From that, the MWT-DSSNL Method achieves enhanced image quality assessment performance when compared to existing works. The simulation of the MWT-DSSNL Method is conducted on factors such as PSNR, average processing time, quality detection accuracy, and false positive rate with different numbers of input images. The simulation result depicts that the MWT-DSSNL method increases the accuracy and also minimizes the time of image quality assessment when compared to conventional works. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024

关键词： Network layers

来源：评论

学校读者我要写书评

暂无评论

Integrating Transformer and Bidirectional Long Short-Term Memory for Intelligent Breast Cancer Detection from Histopathology Biopsy Images

引用

computer Modeling in engineering & sciences 2024年第10期141卷 443-458页

作者： Prasanalakshmi Balaji Omar Alqahtani Sangita Babu Mousmi Ajay Chaurasia Shanmugapriya Prakasam Department of Computer Science College of Computer ScienceKing Khalid UniversityAbha61451Saudi Arabia College of Science and Arts in Rijal Alma King Khalid UniversityAbha61451Saudi Arabia Department of Information Technology Muffakham Jah College of Engineering and TechnologyHyderabad500034India Department of Computer Science and Business Systems Rajalakshmi Engineering CollegeThandalam602105India

Breast cancer is a significant threat to the global population,affecting not only women but also a threat to the entire *** recent advancements in digital pathology,Eosin and hematoxylin images provide enhanced clarity in examiningmicroscopic features of breast tissues based on their staining *** cancer detection facilitates the quickening of the therapeutic process,thereby increasing survival *** analysis made by medical professionals,especially pathologists,is time-consuming and challenging,and there arises a need for automated breast cancer detection *** upcoming artificial intelligence platforms,especially deep learning models,play an important role in image diagnosis and ***,the histopathology biopsy images are taken from standard data ***,the gathered images are given as input to the Multi-Scale Dilated Vision Transformer,where the essential features are ***,the features are subjected to the Bidirectional Long Short-Term Memory(Bi-LSTM)for classifying the breast cancer *** efficacy of the model is evaluated using divergent *** compared with other methods,the proposed work reveals that it offers impressive results for detection.

关键词： Bidirectional long short-term memory breast cancer detection feature extraction histopathology biopsy images multi-scale dilated vision transformer

来源：评论

学校读者我要写书评

暂无评论

Confidence-based Class Relation Embedding for Few-shot Domain Adaptation

Confidence-based Class Relation Embedding for Few-shot Domai...

引用

2024 International Symposium on Internet of Things and Smart Cities, ISITSC 2024

作者： Jia, Baoxi Chen, Junyang Wu, Yirui College of Computer Science and Software Engineering Hohai University Nanjing China College of Computer Science and Software Engineering Shenzhen University Shenzhen China

ISBN: (纸本)9798331505516

In visual tasks such as image classification, the presence of domain shift often renders deep neural network models trained solely on specific datasets unable to generalize to new domains. In practical applications, due to the high cost of annotating rare data, it is extremely challenging to label all source domain samples comprehensively. Moreover, existing unsupervised domain adaptation methods often lack modeling of class relationships. Therefore, we propose a few-shot unsupervised domain adaptation method based on confidence-guided class relationship embedding. This approach aims to identify high-confidence target samples corresponding to sparsely labeled source samples, forming an inter-domain cross-mixed dataset. By constructing inter-domain contrastive learning, we explicitly guide the transfer of knowledge from the source domain to the target domain, leveraging the domain invariance of class relationships. Additionally, considering the varying learning difficulties of different target samples, we have devised an intra-domain class relationship contrastive loss, which leverages easy samples to facilitate the learning of difficult-to-classify samples. Experiments demonstrate that this model contributes to enhancing the performance of domain adaptation tasks. © 2024 IEEE.

关键词： Contrastive 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：