检索结果-内蒙古大学图书馆

5th International Conference on Frontiers technology of Information and computer, ICFTIC 2023

作者： Gao, Yuan Shi, Tianwei Xu, Teng Ding, Jianbang Jing, Rongchen University of Science and Technology Liaoning School of Computer Science and Software Engineering Anshan China

ISBN: (纸本)9798350309034

In recent years, the combination of deep reinforcement learning and unmanned aerial vehicle (UAV) to achieve autonomous flight has been a hot research field. In this paper, an obstacle avoidance navigation algorithm (PA-SAC) based on priority experience buffer pool (PEBP), attention, and Soft-Actor-Critic (SAC) is proposed to solve the continuous space obstacle avoidance navigation problem of UAVs by using a deep reinforcement learning algorithm. In the two simulation experiments, the success rate of the PA-SAC algorithm was 95.6% in the known environment and 73% in the unknown environment. These results demonstrate that the PA-SAC algorithm can achieve autonomous obstacle avoidance for UAVs with deep image input. © 2023 IEEE.

关键词： Attention component Obstacle Avoidance PEBP UAVs

来源：评论

学校读者我要写书评

暂无评论

Pre-trained SAM as data augmentation for image segmentation

引用

CAAI Transactions on Intelligence technology 2025年第1期10卷 268-282页

作者： Junjun Wu Yunbo Rao Shaoning Zeng Bob Zhang Yangtze Delta Region Institute(Huzhou) University of Electronic Science and Technology of ChinaHuzhouChina School of Information and Software Engineering University of Electronic Science and Technology of ChinaChengduChina Pattern Analysis and Machine Intelligence Research Group Department of Computer and Information ScienceUniversity of MacaoMacaoChina

Data augmentation plays an important role in training deep neural model by expanding the size and diversity of the ***,data augmentation mainly involved some simple transformations of ***,in order to increase the diversity and complexity of data,more advanced methods appeared and evolved to sophisticated generative ***,these methods required a mass of computation of training or *** this paper,a novel training-free method that utilises the Pre-Trained Segment Anything Model(SAM)model as a data augmentation tool(PTSAM-DA)is proposed to generate the augmented annotations for *** the need for training,it obtains prompt boxes from the original annotations and then feeds the boxes to the pre-trained SAM to generate diverse and improved *** this way,annotations are augmented more ingenious than simple manipulations without incurring huge computation for training a data augmentation *** comparative experiments on three datasets are conducted,including an in-house dataset,ADE20K and *** this in-house dataset,namely Agricultural Plot Segmentation Dataset,maximum improvements of 3.77%and 8.92%are gained in two mainstream metrics,mIoU and mAcc,***,large vision models like SAM are proven to be promising not only in image segmentation but also in data augmentation.

关键词： data augmentation image segmentation large model segment anything model

来源：评论

学校读者我要写书评

暂无评论

Multi-information Self-attention Autoencoder Sequential recommendation 5

Multi-information Self-attention Autoencoder Sequential reco...

引用

5th International Conference on Frontiers technology of Information and computer, ICFTIC 2023

作者： Wang, Nan Liu, Yang University of Science and Technology Liaoning School of Computer Science and Software Engineering Anshan China

ISBN: (纸本)9798350309034

The field of sequential recommendation plays a crucial role in personalized recommendation systems, aiming to model users' past interactions and predict their future interactions with items or behaviors. Traditional methods in sequential recommendation typically rely on user behavior history and item attributes for making recommendations, but they overlook the internal relationships and contextual information among items within a sequence. Moreover, existing autoencoder models face limitations in capturing long-term dependencies and effectively modeling contextual information for sequential recommendation tasks. To address these issues, we propose a novel framework called miSAASRec that leverages a multi-information autoencoder with a self-attention module to capture internal relevance and contextual features of the data. This enables miSAASRec to achieve more accurate and comprehensive data encoding, thereby enhancing the performance of the autoencoder. We conducted several experiments to demonstrate the superior performance of our miSAASRec model compared to existing methods, achieving improvements ranging from 7.92% and 32.20% in MRR (Mean Reciprocal Rank) and 9.00% and 28.90% in Recall@10. © 2023 IEEE.

关键词： Multi-information autoencoder Self-attention Sequential recommendation

来源：评论

学校读者我要写书评

暂无评论

Multi-Person Estimation Method Combining Interactive and Contextual Information 5

Multi-Person Estimation Method Combining Interactive and Con...

引用

5th International Seminar on Artificial Intelligence, Networking and Information technology, AINIT 2024

作者： Chen, Zhuxiang Li, Shujie Jia, Wei School of Computer Science and Information Engineering Hefei University of Technology Hefei China School of Software Hefei University of Technology Hefei China

ISBN: (纸本)9798350385557

Multi-person pose estimation based on monocular cameras is one of the hot research topics in computer vision. Current monocular multi-person 3D pose estimation methods often treat individuals as independent entities for estimation and are based solely on single-frame assessments. This methodology has notable limitations, particularly its insufficient utilization of the rich interactive information interlinking individuals, as well as the contextual continuity inherent across sequential frames. To address these challenges, our study introduces a novel method for monocular multi-person 3D pose estimation. Firstly, it utilizes the attention mechanism of Transformer to learn the contextual and interactive information of both local and global coordinates. Subsequently, the methd employs low-order convolution operators of a root trajectory network to further assimilate global coordinate information. Finally, an optimization algorithm is used to adaptively filter frames with significant occlusions. Our method effectively integrates the interactive information among multiple individuals and the contextual information across multiple frames. Experimental results on the public dataset MuPoTS-3D demonstrate that compared to the state-of-the-art methods, our approach shows a 4.8% improvement in average joint accuracy and a 19. 4% reduction in average root joint error. © 2024 IEEE.

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

VMCTE:Visualization-BasedMalware Classification Using Transfer and Ensemble Learning

引用

computers, Materials & Continua 2023年第5期75卷 4445-4465页

作者： Zhiguo Chen Jiabing Cao Engineering Research Center of Digital Forensics Ministry of EducationNanjing University of Information Science and TechnologyNanjing210044China School of Computer and Software Nanjing University of Information Science and TechnologyNanjing210044China

The Corona Virus Disease 2019(COVID-19)effect has made telecommuting and remote learning the *** growing number of Internet-connected devices provides cyber attackers with more attack *** development of malware by criminals also incorporates a number of sophisticated obfuscation techniques,making it difficult to classify and detect malware using conventional ***,this paper proposes a novel visualization-based malware classification system using transfer and ensemble learning(VMCTE).VMCTE has a strong anti-interference *** if malware uses obfuscation,fuzzing,encryption,and other techniques to evade detection,it can be accurately classified into its corresponding malware *** traditional dynamic and static analysis techniques,VMCTE does not require either reverse engineering or the aid of domain expert *** proposed classification system combines three strong deep convolutional neural networks(ResNet50,MobilenetV1,and MobilenetV2)as feature extractors,lessens the dimension of the extracted features using principal component analysis,and employs a support vector machine to establish the classification *** semantic representations of malware images can be extracted using various convolutional neural network(CNN)architectures,obtaining higher-quality features than traditional *** fine-tuned and non-fine-tuned classification models based on transfer learning can greatly enhance the capacity to classify various families *** experimental findings on the Malimg dataset demonstrate that VMCTE can attain 99.64%,99.64%,99.66%,and 99.64%accuracy,F1-score,precision,and recall,respectively.

关键词： Malware classification ensemble learning visualization transfer learning

来源：评论

学校读者我要写书评

暂无评论

Double truncated nuclear norm-based matrix decomposition with application to background modeling

引用

Journal of Ambient Intelligence and Humanized Computing 2023年第11期14卷 14921-14930页

作者： Yang, Zhangjing Zhang, Hui Xu, Danhua Zhang, Fanlong Yang, Guowei School of Technology Nanjing Audit University Nanjing211815 China School of Computer and Software Nanjing University of Information Science & Technology Nanjing210044 China

Many topics in pattern recognition and machine learning, such as subspace learning, image restoration, background modeling, can be viewed as the matrix decomposing problem. Double nuclear norm-based matrix decomposition (DNMD) is a new emerging method for dealing with the image data corrupted by continuous occlusion. The method uses a unified low rank assumption to characterize the real image data and continuous occlusion. However, one major limitation of the nuclear norm is that each singular value is treated equally, since the nuclear norm is defined as the sum of all singular values. Thus the rank function may not be well approximated in practice. To overcome this drawback, this paper presents double truncated nuclear norm-based matrix decomposition (DTNMD). The truncated nuclear norm can reflect the rank function more accurate and robust. Experimental results show encouraging results of the proposed algorithm in comparison to the state-of-the-art matrix completion methods on both synthetic and real visual datasets. © 2018, Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Principal component analysis

来源：评论

学校读者我要写书评

暂无评论

Unsupervised feature selection via multiple graph fusion and feature weight learning

引用

science China(Information sciences) 2023年第5期66卷 56-72页

作者： Chang TANG Xiao ZHENG Wei ZHANG Xinwang LIU Xinzhong ZHU En ZHU School of Computer Science China University of Geosciences State Key Laboratory for Novel Software Technology Nanjing University School of Computer National University of Defense Technology National Supercomputing Center in Jinan Qilu University of Technology (Shandong Academy of Sciences) College of Mathematics Physics and Information Engineering Zhejiang Normal University

Unsupervised feature selection attempts to select a small number of discriminative features from original high-dimensional data and preserve the intrinsic data structure without using data labels. As an unsupervised learning task, most previous methods often use a coefficient matrix for feature reconstruction or feature projection, and a certain similarity graph is widely utilized to regularize the intrinsic structure preservation of original data in a new feature space. However, a similarity graph with poor quality could inevitably afect the final results. In addition, designing a rational and efective feature reconstruction/projection model is not easy. In this paper, we introduce a novel and efective unsupervised feature selection method via multiple graph fusion and feature weight learning（MGF2WL） to address these issues. Instead of learning the feature coefficient matrix, we directly learn the weights of diferent feature dimensions by introducing a feature weight matrix, and the weighted features are projected into the label space. Aiming to exploit sufficient relation of data samples, we develop a graph fusion term to fuse multiple predefined similarity graphs for learning a unified similarity graph, which is then deployed to regularize the local data structure of original data in a projected label space. Finally, we design a block coordinate descent algorithm with a convergence guarantee to solve the resulting optimization problem. Extensive experiments with sufficient analyses on various datasets are conducted to validate the efficacy of our proposed MGF2WL.

关键词： unsupervised feature selection graph fusion feature weight learning local structure preservation

来源：评论

学校读者我要写书评

暂无评论

Deepfake Detection: Emerging Techniques and Evolving Challenges 12

Deepfake Detection: Emerging Techniques and Evolving Challen...

引用

12th International Conference on Internet of Everything, Microwave, Embedded, Communication and Networks, IEMECON 2024

作者： Natarajan, Krishnaraj Mathur, Abeer Bhargava, Kshitiz Singh, Manvendra Tejpal, Moulik Department of Database Systems School of Computer Science & Engineering Vellore Institute of Technology Vellore India Department of Software Systems School of Computer Science & Engineering Vellore Institute of Technology Vellore India

ISBN: (纸本)9798350387315

Several newly developed techniques and tools for manipulating images, audio, and videos have been introduced as an outcome of the recent and rapid breakthroughs in AI, machine learning, and deep learning. While most applications for these techniques or tools are in the fields of entertainment and education, some individuals with unlawful intent have also been benefited from them. These individuals use such techniques for various purposes, including the spread of misleading information and unnecessary propaganda, the incitement of political instability, hate and unrest, as well as for purposes of torture and blackmail. These high-quality and convincing manipulated images, audio, or videos are commonly referred to as ‘Deepfakes’. Since then, various solutions to the problems raised by Deepfakes have been proposed in academic studies. This literature review contains relevant publications that offered a variety of approaches to give an updated summary of the research activities in different types of Deepfake attacks, their detection, and countermeasures. It also assesses the effectiveness of the detection capabilities of different techniques with various datasets and algorithms applied in Deepfake detection, while also outlining the various benefits and drawbacks of various methodologies. © 2024 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Improved Proximal Policy Optimization Algorithm for Controller Design in Hybrid UAVs∗

Improved Proximal Policy Optimization Algorithm for Controll...

引用

2024 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2024

作者： Qi, Mingyu Zheng, Hongyuan Zhai, Xiangping Bryce Zhu, Qi School of Computer Science and Technology School of Artificial Intelligence School of Software at Nanjing University of Aeronautics and Astronautics Nanjing211106 China

ISBN: (纸本)9781665410205

Drones have become an indispensable tool in our daily lives. While fixed-wing and rotary-wing drones are common types, each comes with its own set of advantages and disadvantages. Hybrid drones, however, combine the strengths of both types, enabling vertical takeoff and landing, hovering, and remote flight. Nonetheless, the aerodynamics of hybrid drones are exceedingly intricate, leading to a sluggish pace in their development. In this article, we propose a neural network controller design, employing the reinforcement learning Proximal Policy Optimization (PPO) algorithm to train the controller. Additionally, we integrate an attention mechanism into the network's input section to emphasize the speed variable, thereby enhancing the data processing and improving the model performance and efficiency. Experimental results demonstrate that our approach yields a controller with superior stability and optimal value. © 2024 IEEE.

关键词： Drones

来源：评论

学校读者我要写书评

暂无评论

Unified Approach Integrating Machine Learning Algorithms for Detection of e-Mail Phishing Attacks 1

Unified Approach Integrating Machine Learning Algorithms for...

引用

1st International Conference on Computing, sciences and Communications, ICCSC 2024

作者： Mathur, Abeer Dubey, Abhinav Sudhakar Mahara, Pranav Singh Gandhi, Usha Devi School of Computer Science & Engineering Vellore Institute of Technology Department of Software Systems Vellore India School of Computer Science Engineering & Information Systems Vellore Institute of Technology Department of Information Technology Vellore India

ISBN: (纸本)9798350353648

With the ever-rising risk of phishing attacks, which capitalize on vulnerable human behavior in the contemporary digital space, requires new cybersecurity methods. This literary work contributes to the solution by novel incorporation of three techniques: Support Vector Machine (SVM), Natural Language Processing (NLP) and Random Forest (RF). Utilizing NLP for deep analysis of text in phishing e-mails, SVM for strong categorization and Random Forest's decision making by group helps in improving detection. This combination of language and pattern analysis along with group learning builds a fortified system. This approach differs from prior work on a variety of factors such as reliability, security, integrity, scalability, interpretability, transparency, and robustness. Thorough evaluation on a reliable dataset aimed to demonstrate the method's effectiveness, which resulted in 98.7 % of accuracy in detection. This majorly aims at showcasing the potential of ensemble techniques to combat against cyber-attacks like phishing for laying a strong foundation for a secure digital environment. © 2024 IEEE.

关键词： Phishing

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：