检索结果-内蒙古大学图书馆

Gender and age-evolution detection based on audio forensic analysis using light deep neural network

International Journal of Speech technology 2023年第4期26卷 1091-1098页

作者： AL-Shakarchy, Noor D. Rageb, Huda Safoq, Mais Saad Department of Computer Science Faculty of Computer Science and Information Technology Kerbala University Kerbala Iraq

Forensic audio analysis is a foundation stone of many crime investigations. In forensic evidence;the audio file of the human voice is analyzed to extract much information in addition to the content of the speech, such as the speaker’s identity, emotions, gender, origin, etc. The accurate determination of individuals into groups based on their age development stage and their gender are often used as early investigations to differentiate them and determine the legal rights and responsibilities associated with them. This work introduces a light CNN model with a new architecture to detect the human age-evolution being’s stage (kids or adults) at the same time the gender of the adult one (male or female) based on the individual’s voice characteristics, which offers a balance between computational efficiency and model accuracy. The temporal information in the audio file is prepared by scaled and normalized. Then this information is exploited to extract and track the unique and salient audio features that make up the pattern of the feature map for each target class through some convolutional layers followed by maxpooling layers. Finally, The decision is made based on these feature maps by some fully connected layers. Successful and promising results are accomplished in terms of accuracy and loss functions which realize 0.99 and 0.017 respectively over the riched Voxceleb2 dataset. The proposed model underscores the importance of leveraging Light DNNs for gender and age-evolution detection, offering a robust and ethically sound solution for real-world applications in the field of audio forensics such as span speaker identification, victim profiling, deception detection, and more, contributing to the advancement of audio forensic analysis. © 2023, The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature.

关键词： Computational efficiency

来源：评论

学校读者我要写书评

暂无评论

Enhancing sustainable 5G powered agriculture 4.0 : Summary of low power connectivity, internet of UAV things, AI solutions and research trends

引用

Multimedia Tools and Applications 2025年第17期84卷 17389-17433页

作者： Majumdar, Parijata Mitra, Sanjoy Bhattacharya, Diptendu Bhushan, Bharat Department of Computer Science and Engineering Techno College of Engineering Agartala Maheshkhola Madhuban799004 India Department of Computer Science and Engineering Tripura Institute of Technology Agartala Narsingarh Agartala799009 India Department of Computer Science and Engineering National Institute of Technology Agartala Jirania Agartala799046 India Department of Computer Science and Engineering School of Engineering and Technology Sharda University Greater Noida201310 India

Emerging technologies of Agriculture 4.0 such as the Internet of Things (IoT), Cloud Computing, Artificial Intelligence (AI), and 5G network services are being rapidly deployed to address smart farming implementation-related concerns and to fulfill the increasing food demand of the expanding demographics. Adoption of 5G offers numerous benefits including Ultra-Reliable Low Latency Communication (URLLC) required to reduce communication delays in time-sensitive applications, non-public networks to allocate part of the frequency spectrum on demand, and network slicing to fulfill the diverse requirements requested by a specific application etc. for realizing the full potential of Agriculture 4.0 operations. In this paper, a comprehensive review of 5G radio enhancements, enabling technologies, and multi-tier heterogeneous 5G network architecture are presented. The main focus is to concisely cover AI integration with Unmanned Aerial Vehicles (UAVs) leveraging 5G technologies and Low Power Wide Area Networks (LPWANs) for supporting intelligent, long-range, cost-effective and energy-saving data transmission across the network and seamless Agriculture 4.0 operations. Cloud computing architectures that permit centralized management of Agriculture 4.0 operations through sharing of computations, configurable networks, and servers are also studied in detail to reveal the need for edge computation as increasing number of IoT devices allows only to perform complex computations on-site. Extensive survey reveals the recent advances of 5G-enabled Agriculture 4.0 operations, future research opportunities of 5G networks including Energy Harvesting techniques for the deployment of massive to critical IoT applications, and the socio-demographic impacts of 5G networks in Agriculture 4.0 to support ubiquitous connectivity of exponentially increasing IoT devices. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Energy efficiency

来源：评论

学校读者我要写书评

暂无评论

Inductive Lottery Ticket Learning for Graph Neural Networks

引用

Journal of computer science & technology 2024年第6期39卷 1223-1237页

作者： Yong-Duo Sui Xiang Wang Tianlong Chen Meng Wang Xiang-Nan He Tat-Seng Chua School of Data Science University of Science and Technology of ChinaHefei 230027China Department of Electrical and Computer Engineering The University of Texas at AustinAustin 78712U.S.A School of Computer Science and Information Engineering Hefei University of TechnologyHefei 230009China School of Computing National University of SingaporeSingapore

Graph neural networks (GNNs) have gained increasing popularity, while usually suffering from unaffordable computations for real-world large-scale applications. Hence, pruning GNNs is of great need but largely unexplored. The recent work Unified GNN Sparsification (UGS) studies lottery ticket learning for GNNs, aiming to find a subset of model parameters and graph structures that can best maintain the GNN performance. However, it is tailed for the transductive setting, failing to generalize to unseen graphs, which are common in inductive tasks like graph classification. In this work, we propose a simple and effective learning paradigm, Inductive Co-Pruning of GNNs (ICPG), to endow graph lottery tickets with inductive pruning capacity. To prune the input graphs, we design a predictive model to generate importance scores for each edge based on the input. To prune the model parameters, it views the weight’s magnitude as their importance scores. Then we design an iterative co-pruning strategy to trim the graph edges and GNN weights based on their importance scores. Although it might be strikingly simple, ICPG surpasses the existing pruning method and can be universally applicable in both inductive and transductive learning settings. On 10 graph-classification and two node-classification benchmarks, ICPG achieves the same performance level with 14.26%–43.12% sparsity for graphs and 48.80%–91.41% sparsity for the GNN model.

关键词： lottery ticket hypothesis graph neural networks neural network pruning

来源：评论

学校读者我要写书评

暂无评论

Video captioning – a survey

引用

Multimedia Tools and Applications 2025年第2期84卷 947-978页

作者： Vaishnavi, J. Narmatha, V. Department of Computer Applications SRM Institute of Science and Technology Trichy India Department of Computer and Information Science Annamalai University Annamalai Nagar Tamilnadu 608002 India

The explosion of the novel phenomenon of the combination of computer vision and Natural language processing is playing a vital role in converting the ordinary world into a more technological pool. Natural language processing and computer vision are the vanguard of Artificial intelligence with the enormous potential currently ruling more fields. The trending Accelerated research of the NLP and CV combo are cancer screening, surgical simulation, visual properties description, visual retrieval, visual description, etc. Video captioning is a task of localizing the event and various features are extracted from the particular event and then the description of the entire event is performed with Natural languages using various techniques. The task of video captioning has some evolution of various techniques from Traditional methods to Deep learning. Currently, Deep learning techniques are ruling the field. Several models are being proposed for the better enhancement of the video captioning task with the default Encoder-Decoder framework and some models with the variance of Attention mechanism and Transfer learning etc., The Attainment of every technique and the accuracy of the generated results are highly dependent on the nature of the problem, diversity, performance of chosen dataset and construction of the technique with various layers, and the amount of data split for train, validate, and test. This survey discusses the captioning types, real-time applications of video captioning, evolution of various methods in video captioning, advanced techniques, different datasets, metrics that are used to evaluate the results, and analysis of the results of the existing models for video captioning. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

Comprehensive Analysis of VM Migration Trends in Cloud Data Centers

引用

Recent Patents on Engineering 2023年第6期17卷 134-146页

作者： Singh, Shalu Singh, Dinesh Department of Computer Science & Technology Manav Rachna University Faridabad India Department of Computer Science & Engineering Deenbandhu Chhotu Ram University of Science and Technology Sonepat Murthal India

Background: Virtualization adequately maintains increasing requirements for storage, networking, servers, and computing in exhaustive cloud data centers (CDC)s. Virtualization assists in gaining different objectives like dedicated server sustenance, fault tolerance, comprehensive service availability, and load balancing, by virtual machine (VM) migration. The VM migration process continuously requires CPU cycles, communication bandwidth, memory, and processing power. Therefore, it detrimentally prevails over the performance of dynamic applications and cannot be completely neglected in the synchronous large-scale CDC, explicitly when service level agreement (SLA) and analytical trade goals are to be defined. Introduction: Live VM migration is intermittently adopted as it grants the operational service even when the migration is executed. Currently, power competence has been identified as the primary design requirement for the current CDC model. It grows from a single server to numerous data centres and clouds, which consume an extensive amount of electricity. Consequently, appropriate energy management techniques are especially important for CDCs. Methods: This review paper delineates the need for energy efficiency in the CDC, the systematic mapping of VM migration methods, and research pertinent to it. After that, an analysis of VM migration techniques, the category of VM migration, duplication, and context-based VM migration is presented along with its relative analysis. Results: The various VM migration techniques were compared on the basis of various performance measures. The techniques based on duplication and context-based VM migration methods provide an average reduction in migration time of up to 38.47%, data transfer rate of up to 51.4%, migration downtime of up to 36.33%, network traffic rate of up to 44% and reduced application efficiency overhead up to 14.27%. Conclusion: The study aids in analyzing threats and research challenges related to VM migration

关键词： Virtual machine

来源：评论

学校读者我要写书评

暂无评论

Faster AMEDA-A Hybrid Mesoscale Eddy Detection Algorithm

引用

computer Modeling in Engineering & sciences 2024年第11期141卷 1827-1846页

作者： Xinchang Zhang Xiaokang Pan Rongjie Zhu Runda Guan Zhongfeng Qiu Biao Song Organization Department Nanjing University of Information Science and TechnologyNanjing210044China School of Computer Science Nanjing University of Information Science and TechnologyNanjing210044China School of Teacher Education Nanjing University of Information Science and TechnologyNanjing210044China School of Electronics&Information Engineering Nanjing University of Information Science and TechnologyNanjing210044China School of Software Nanjing University of Information Science and TechnologyNanjing210044China

Identification of ocean eddies from a large amount of ocean data provided by satellite measurements and numerical simulations is crucial,while the academia has invented many traditional physical methods with accurate detection capability,but their detection computational efficiency is *** recent years,with the increasing application of deep learning in ocean feature detection,many deep learning-based eddy detection models have been developed for more effective eddy detection from ocean *** it is difficult for them to precisely fit some physical features implicit in traditional methods,leading to inaccurate identification of ocean *** this study,to address the low efficiency of traditional physical methods and the low detection accuracy of deep learning models,we propose a solution that combines the target detection model Faster Region with CNN feature(Faster R-CNN)with the traditional dynamic algorithm Angular Momentum Eddy Detection and Tracking Algorithm(AMEDA).We use Faster R-CNN to detect and generate bounding boxes for eddies,allowing AMEDA to detect the eddy center within these bounding boxes,thus reducing the complexity of center *** demonstrate the detection efficiency and accuracy of this model,this paper compares the experimental results with AMEDA and the deep learning-based eddy detection method *** results show that the eddy detection results of this paper are more accurate than eddyNet and have higher execution efficiency than AMEDA.

关键词： Mesoscale eddy detection object detection contour detection remote sensing

来源：评论

学校读者我要写书评

暂无评论

A level set model with shape prior constraint for intervertebral disc MRI image segmentation

引用

Multimedia Tools and Applications 2025年第9期84卷 6755-6783页

作者： Tian, Zhuangzhou Wang, Shengsheng College of Computer Science and Technology Jilin University Changchun130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun130012 China

Accurate intervertebral disc image segmentation is necessary for further treatment. However, existing methods are difficult to segment due to the intensity inhomogeneity of intervertebral disc MRI images and the similarity in intensity between the intervertebral disc and the surrounding regions. This paper proposes a novel level set model by introducing a shape prior constraint for automatic and accurate segmentation. We utilize a detection model and intensity change to obtain rectangle prior information, which is used to define a shape constraint term. Then, combining data and regularizing terms construct an energy function and employ the steepest descent method minimization to get segmentation results. Finally, we apply the proposed new method to segment intervertebral disc images and get promising results. Extensive experiments and comparisons show that our model is more efficient and accurate. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Bridging the gap in biomedical information retrieval: Harnessing machine learning for enhanced search results and query semantics

引用

Journal of Intelligent and Fuzzy Systems 2024年第4期46卷 9491-9510页

作者： Madhubala, P. Ghanimi, Hayder M.A. Sengan, Sudhakar Abhishek, Kumar Department of Computer Science and Engineering Bharathiyar Institute of Engineering for Women Tamil Nadu Salem India Information Technology Department College of Science University of Warith Al-Anbiyaa Karbala Iraq Computer Science Department College of Computer Science and Information Technology University of Kerbala Karbala Iraq Department of Computer Science and Engineering PSN College of Engineering and Technology Tamil Nadu Tirunelveli India Department of Mathematics VIT Bhopal Bhopal India

The medical domain faces unique challenges in Information Retrieval (IR) due to the complexity of medical language and terminology discrepancies between user queries and documents. While traditional Keyword-Based Methods (KBM) have limitations, the integration of semantic knowledge bases and concept mapping techniques enhances data organization and retrieval. Addressing the growing demands in the biomedical field, a novel medical Information Retrieval System (IRS) is proposed that employs Deep Learning (DL) and KBM. This system comprises five core steps: pre-processing of texts, document indexing using DL (ELMo) and KBM, advanced query processing, a BiLSTM-based retrieval network for contextual representation, and a KR-R re-ranking algorithm to refine document relevance. The purpose of the system is to give users improved biomedical search results through the integration of all of these techniques into a method that takes into consideration the semantic problems of medical records. An in-depth examination of the TREC-PM track samples from 2017 to 2019 observed an impressive leading MRR score of 0.605 in 2017 and a best-in-class rPrec score of 0.350 in 2019, proving how well able the system is to detect and rank relevant medical records accurately. © 2024 – IOS Press. All rights reserved.

关键词： Information retrieval

来源：评论

学校读者我要写书评

暂无评论

Facile Synthesis of Polyisothioureas via Alternating Copolymerization of Aziridines and Isothiocayanates

引用

Chinese Journal of Polymer science 2024年第3期42卷 292-298,I0005页

作者： Hao-Tian Zhang Ming-Xin Niu Qi Zhang Chen-Yang Hu Xuan Pang Key Laboratory of Polymer Ecomaterials Changchun Institute of Applied ChemistryChinese Academy of SciencesChangchun 130022China Department of Materials Science and Engineering Jilin UniversityChangchun 130025China School of Applied Chemistry and Engineering University of Science and Technology of ChinaHefei 230026China

Isothiourea is an important class of sulfur-containing molecules showing unique catalytic and biological activities. As such,polyisothiourea is envisioned to be an interesting type of polymer that potentially exhibits a number of interesting properties. However, there is no access to synthesizing well-defined polyisothiourea, and currently isothiourea-containing polymers are mainly prepared by immobilizing onto other polymer's side chain. Herein, we report the first facile synthesis of polyisothioureas via alternating copolymerization of aziridines and isothiocayanates. Mediated by the catalytic system of phosphazene superbases/alcohol, a broad scope of aziridines and isothiocayanates could be transformed into polyisothioureas with adjustable substitutions(11 examples). The structures of obtained polyisothioureas were fully characterized with ^(1)H-NMR, ^(13)C-NMR, and ^(1)H-^(13)C HMBC NMR. Moreover, the polyisothioureas show tunable thermal properties depending on substitutions on the isothiourea linkages. The novel structure of these polyisothioureas will enable a powerful platform for the discovery of nextgeneration functional plastics.

关键词： Polyisothioureas Alternative copolymer Aziridines Isothiocayanates

来源：评论

学校读者我要写书评

暂无评论

AGG: attention-based gated convolutional GAN with prior guidance for image inpainting

引用

Neural Computing and Applications 2024年第20期36卷 12589-12604页

作者： Yu, Xiankang Dai, Lei Chen, Zhihua Sheng, Bin Department of Computer Science and Engineering East China University of Science and Technology Shanghai200237 China Department of Computer Science and Engineering Shanghai Jiao Tong University Shanghai200240 China

Image inpainting has made great achievements recently, but it is often tough to generate a semantically consistent image when faced with large missing areas in complex scenes. To address semantic and structural alignment in existing methods for image inpainting, this paper proposes an end-to-end attention-based gated convolution GAN with prior guidance named AGG, which designs the spatial and channel attention mechanisms for full extraction of semantic and structural features. Moreover, AGG constructs the attention-based upsampling module based on the channel attention to refine the feature map and capture more details from features of up-level low sizes. AGG uses the image contour as prior, allowing the gated convolution and attention mechanism may fill the image efficiently by focusing on the contour information. The attention-based gated convolution can effectively capture the global features and compensate for the limitations of the restricted receptive field of the naive convolution. Compared to other models, AGG generates images with finer outline features and no common problems such as the watermark and blur, which shows the best overall performance on the Paris StreetView, CelebA-HQ, and Places2 datasets. The best FID, LPIPS, PSNR, and SSIM values achieved by AGG are 2.18, 0.046, 30.82, and 0.951 on the CelebA-HQ dataset, with at least 3.21% and 6.52% performance improvement on FID and LPIPS compared to state-of-the-art methods, respectively. The source code will be available at https://***/Shawn-Yu-1/AGGNet. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024. corrected publication 2024.

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：