检索结果-内蒙古大学图书馆

Multimodal Cross-Attention Mechanism-based Algorithm for Elderly Behavior Monitoring and Recognition

Chinese Journal of Electronics 2025年第1期34卷 309-321页

作者： Hao Liu Zhiquan Feng Qingbei Guo School of Information Science and Engineering Unicersity of Jinan Shandong Provincial Key Laboratory of Network Based Intelligent Computing Unicersity of Jinan

In contrast to the general population, behavior recognition among the elderly poses increased specificity and difficulty, rendering the reliability and usability aspects of safety monitoring systems for the elderly more challenging. Hence, this study proposes a multi-modal perception-based solution for an elderly safety monitoring recognition system. The proposed approach introduces a recognition algorithm based on multi-modal cross-attention mechanism, innovatively incorporating complex information such as scene context and voice to achieve more accurate behavior recognition. By fusing four modalities, namely image, skeleton, sensor data, and audio, we further enhance the accuracy of recognition. Additionally, we introduce a novel human-robot interaction mode, where the system associates directly recognized intentions with robotic actions without explicit commands, delivering a more natural and efficient elderly assistance paradigm. This mode not only elevates the level of safety monitoring for the elderly but also facilitates a more natural and efficient caregiving approach. Experimental results demonstrate significant improvement in recognition accuracy for 11 typical elderly behaviors compared to existing methods.

关键词： Privacy Accuracy Image recognition Speech recognition Robot sensing systems Skeleton Safety Reliability Older adults Monitoring

来源：评论

学校读者我要写书评

暂无评论

Multi-View Transfer Clustering Algorithm based on Gaussian Mixture Model

Multi-View Transfer Clustering Algorithm Based on Gaussian M...

引用

2024 International Conference on Fuzzy Theory and Its Applications, iFUZZY 2024

作者： Jiang, Yan Li, Xingchen Zhou, Jin Han, Shiyuan Du, Tao Yang, Cheng Liu, Bowen University of Jinan Shandong Provincial Key Laboratory of Network Based Intelligent Computing Jinan250022 China

ISBN: (纸本)9798350352788

It is well known that the Gaussian mixture model (GMM) is an exemplary model-based clustering approach that captures data distribution patterns to facilitate clustering. However, when it faces to multi-view clustering tasks, the classical GMM algorithm often encounters limitations in scalability. Multi-view clustering entails employing diverse views or feature sets to cluster data, thereby achieving comprehensive and accurate clustering outcomes. The classical GMM algorithm fails to fully exploit the correlated information across multiple views, resulting in a decrease in clustering accuracy. This paper proposes a new GMM-based trans-fer multi-view clustering algorithm named MT-GMM. This algorithm transfers valuable knowledge from other views to guide and enhance the clustering task of the current view, thereby improving the clustering performance specifically for the current view. Experimental results conducted on real-world datasets demonstrate that the proposed MT-GMM outperforms several other multi-view clustering methods, as it can effectively harness the complementary information between different views accurately capture the structures of data. © 2024 IEEE.

关键词： Clustering algorithms

来源：评论

学校读者我要写书评

暂无评论

Selective Depth Attention networks for Adaptive Multiscale Feature Representation

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第10期5卷 5064-5074页

作者： Guo, Qingbei Wu, Xiao-Jun Xu, Tianyang Si, Tongzhen Hu, Cong Tian, Jinglan University of Jinan Shandong Provincial Key Laboratory of Network-based Intelligent Computing Jinan250022 China Jiangnan University School of Artificial Intelligence and Computer Science Wuxi214122 China

Existing multiscale methods lead to a risk of just increasing the receptive field sizes while neglecting small receptive fields. Thus, it is a challenging problem to effectively construct adaptive neural networks for recognizing various spatial-scale objects. To tackle this issue, we first introduce a new attention dimension, i.e., depth, in addition to existing attentions such as channel-attention, spatial-attention, branch-attention, and self-attention. We present a novel selective depth attention network to treat multiscale objects symmetrically in various vision tasks. Specifically, the blocks within each stage of neural networks, including convolutional neural networks (CNNs), e.g., ResNet, SENet, and Res2Net, and vision transformers (ViTs), e.g., PVTv2, output the hierarchical feature maps with the same resolution but different receptive field sizes. based on this structural property, we design a depthwise building module, namely an selective depth attention (SDA) module, including a trunk branch and a SE-like attention branch. The block outputs of the trunk branch are fused to guide their depth attention allocation through the attention branch globally. According to the proposed attention mechanism, we dynamically select different depth features, which contributes to adaptively adjusting the receptive field sizes for the variable-sized input objects. Moreover, our method is orthogonal to multiscale networks and attention networks, so-called SDA-xNet. Extensive experiments demonstrate that the proposed SDA method significantly improves the original performance as a lightweight and efficient plug-in on numerous computer vision tasks, e.g., image classification, object detection, and instance segmentation. © 2020 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Applications and potentials of machine learning in optoelectronic materials research:An overview and perspectives

引用

Chinese Physics B 2023年第12期32卷 108-128页

作者：张城洲付小倩 School of Information Science and Engineering University of JinanJinan 250022China Shandong Provincial Key Laboratory of Network based Intelligent Computing University of JinanJinan 250022China

Optoelectronic materials are essential for today's scientific and technological development,and machine learning provides new ideas and tools for their *** this paper,we first summarize the development history of optoelectronic materials and how materials informatics drives the innovation and progress of optoelectronic materials and ***,we introduce the development of machine learning and its general process in optoelectronic materials and describe the specific implementation *** focus on the cases of machine learning in several application scenarios of optoelectronic materials and devices,including the methods related to crystal structure,properties(defects,electronic structure)research,materials and devices optimization,material characterization,and process *** summarizing the algorithms and feature representations used in different studies,it is noted that prior knowledge can improve optoelectronic materials design,research,and decision-making ***,the prospect of machine learning applications in optoelectronic materials is discussed,along with current challenges and future *** paper comprehensively describes the application value of machine learning in optoelectronic materials research and aims to provide reference and guidance for the continuous development of this field.

关键词： optoelectronic materials devices machine learning prior knowledge

来源：评论

学校读者我要写书评

暂无评论

Design of a new kind of chemical experiment container with virtual reality fusion

引用

Virtual Reality & intelligent Hardware 2023年第4期5卷 317-337页

作者： Lurong YANG Zhiquan FENG Junhong MENG Department of Information Science and Engineering University of JinanJinan 250022China Shandong Provincial Key Laboratory of Network Based Intelligent Computing Jinan 250022China

Background At present,the teaching of experiments in primary and secondary schools is affected by cost and security *** research on virtual experiment platforms has alleviated these ***,the lack of real experimental equipment and use of a single channel to understand user intentions weaken these platforms operationally and degrade the naturalness of *** To solve these problems,we propose an intelligent experimental container structure and a situational awareness algorithm,both of which are verified and applied to a chemical experiment involving virtual-real ***,the acquired images are denoised in the visual channel using the maximum diffuse reflection chroma to remove ***,container situational awareness is realized by segmenting the image liquid level and establishing a relation-fitting ***,strategies for constructing complete behaviors and making priority comparisons among behaviors are adopted for information complementarity and independence,respectively.A multichannel intentional understanding model and an inter-active paradigm that integrates vision,hearing,and touch are *** The experimental results show that the accuracy of the intelligent container situation awareness proposed in this paper reaches 99%,and the accuracy of the proposed intention understanding algorithm reaches 94.7%.The test shows that the intelligent experimental system based on the new interaction paradigm also has better performance and a more realistic sense of operation experience in terms of experimental *** The results indicate that the proposed experimental container and algorithm can achieve a natural level of human-computer interaction in a virtual chemical experiment platform,enhance the user′s sense of operation,and achieve high levels of user satisfaction.

关键词： Multichannel intent fusion understanding Virtual-real fusion Natural interaction Remove image highlights Nonlinear curve fitting

来源：评论

学校读者我要写书评

暂无评论

Deep-reinforcement-learning-based robot motion strategies for grabbing objects from human hands

引用

Virtual Reality & intelligent Hardware 2023年第5期5卷 407-421页

作者： Zeyuan CAI Zhiquan FENG Liran ZHOU Xiaohui YANG Tao XU School of Information Science and Engineering University of JinanJinan 250022China Shandong Provincial Key Laboratory of Network Based Intelligent Computing Jinan 250022China

Background Robot grasping encompasses a wide range of research areas;however, most studies have been focused on the grasping of only stationary objects in a scene;only a few studies on how to grasp objects from a user's hand have been conducted. In this paper, a robot grasping algorithm based on deep reinforcement learning (RGRL) is proposed. Methods The RGRL takes the relative positions of the robot and the object in a user's hand as input and outputs the best action of the robot in the current state. Thus, the proposed algorithm realizes the functions of autonomous path planning and grasping objects safely from the hands of users. A new method for improving the safety of human-robot cooperation is explored. To solve the problems of a low utilization rate and slow convergence of reinforcement learning algorithms, the RGRL is first trained in a simulation scene, and then, the model para-meters are applied to a real scene. To reduce the difference between the simulated and real scenes, domain randomization is applied to randomly change the positions and angles of objects in the simulated scenes at regular intervals, thereby improving the diversity of the training samples and robustness of the algorithm. Results The RGRL's effectiveness and accuracy are verified by evaluating it on both simulated and real scenes, and the results show that the RGRL can achieve an accuracy of more than 80% in both cases. Conclusions RGRL is a robot grasping algorithm that employs domain randomization and deep reinforcement learning for effective grasping in simulated and real scenes. However, it lacks flexibility in adapting to different grasping poses, prompting future research in achieving safe grasping for diverse user postures.

关键词： Robot grasping Deep reinforcement learning Domain randomization Human-to-robot handovers Human-machine collaboration

来源：评论

学校读者我要写书评

暂无评论

Non-Destructive Judgment of Watermelon Ripeness based on Multi-Feature Fusion of Image 20

Non-Destructive Judgment of Watermelon Ripeness Based on Mul...

引用

20th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery, ICNC-FSKD 2024

作者： Shi, Hongbao Liu, Jun Huang, Yimei Li, Jinping School of Information Science and Engineering University of Jinan Shandong Provincial Key Laboratory of Network Based Intelligent Computing Jinan China

ISBN: (纸本)9798350356328

The nondestructive judgment of watermelon ripeness is a key issue in everyday life. Whether watermelons are ripe or not affects the interests of watermelon farmers and consumers. So far, acoustic, electrical, machine vision and other methods have been applied to the nondestructive judgment of agricultural product ripeness, but the equipment used in the lab, such as spectrometers and laser Doppler vibrometers, is expensive, difficult to carry, and has limited practical value in daily life. based on practical experience, the ripeness of watermelons can be judged by observing specific physical characteristics, such as the rind's glossiness, texture, fuzz on the fruit stalk and navel. In this study, images of the watermelon's texture, fuzz and navel were captured using a smartphone. We utilized ResN et with an embedded Coordinate Attention mechanism and Swin-Transformer networks to independently analyze the ripeness of each feature. Then we combined the output prediction scores using post-fusion technology, with the highest weighted prediction score judging the final ripeness assessment. Our method achieved a 90.2% accuracy rate on the test set. The experiments demonstrate that a deep learning model utilizing multi-feature fusion of image data can reliably and accurately identify watermelon ripeness, even under varying image conditions. © 2024 IEEE.

关键词： Laser Doppler velocimeters

来源：评论

学校读者我要写书评

暂无评论

Multi-Information Filter Encoding network for Multi-Label Text Classification 26

Multi-Information Filter Encoding Network for Multi-Label Te...

引用

26th International Conference on Computer Supported Cooperative Work in Design, CSCWD 2023

作者： Li, Leping Ma, Kun Cui, Benkuan University of Jinan Shandong Provincial Key Laboratory of Network Based Intelligent Computing Jinan250022 China

ISBN: (纸本)9798350331684

Multi-label text classification (MLTC) is essential in natural language processing. For real applications, MLTC is challenging when class distribution is long-tailed, that is, a few of labels (head labels) have many documents while most labels (tail labels) have a few documents. In this paper, we propose a Multi-Information Filter Encoding network (MIFEN) to address the issue of long-tailed distribution. Specifically, MIFEN constructs Multi-Information Filter Encoder to filter useless information and optimize the text space and label space. Then, MIFEN enriches the tail label-related features by extracting filtered text-specific label information for each document and highlights significant features of the corresponding text and label information to generate the final document representation. Experiment results have demonstrated that our MIFEN outperforms better than the state-of-the-art methods. © 2023 IEEE.

关键词： Information filtering

来源：评论

学校读者我要写书评

暂无评论

Cancelable fingerprint template protection based on random quantization and improved bloom filter

引用

Multimedia Tools and Applications 2024年 1-33页

作者： Khan, Md Sabuj Li, Hengjian Sun, Yanan Zhao, Chuan Shandong Provincial Key Laboratory of Network Based Intelligent Computing University of Jinan Jinan250022 China School of Information Science and Engineering University of Jinan Jinan250022 China Quan Cheng Laboratory Jinan Jinan250103 China

Improving privacy and security in biometric systems requires addressing vulnerabilities in existing cancelable fingerprint templates based on bloom filters. Therefore, we propose cancelable fingerprint template protection based on random quantization and improved bloom filter. Firstly, random quantization entails subtracting a random key from the original fingerprint vector to generate binary features, ensuring security, and enhancing efficiency. Secondly, the permutation template mechanism rearranges binary features using XOR operations with a security key, enhancing security by concealing original information and incorporating additional features. Finally, the improved bloom filter scheme ensures robust security for cancelable template protection by integrating sub-block binary conversion and Logistic-Tent-Sine (LTTS) random key techniques, guaranteeing the irreversible transformation of the original fingerprint template while maintaining high security. The experimental results demonstrate the scheme's outstanding accuracy on the FVC2002 and FVC2004 fingerprint databases, establishing it as a state-of-the-art solution in fingerprint recognition. Comprehensive security analysis confirms compliance with stringent cancelable biometric protection criteria, ensuring superior irreversibility, unlinkability, revocability, and privacy against various attacks. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Biometrics

来源：评论

学校读者我要写书评

暂无评论

Optimizing Resource Allocation in Edge computing to Reduce Delay based on Multi-Layer Particle Swarm Optimization

Optimizing Resource Allocation in Edge Computing to Reduce D...

引用

2023 International Conference on networks, Communications and intelligent computing, NCIC 2023

作者： Kong, Xiaoman Yang, Bo Han, Yamin University of Jinan Shandong Provincial Key Laboratory of Network-Based Intelligent Computing Jinan250022 China Computing University of Jinan Shandong Provincial Key Laboratory of Network-Based Intelligent Jinan250022 China Quan Cheng Laboratory Shandong Jinan250103 China

ISBN: (纸本)9798350371949

With the popularity of Internet of Things (IoT) applications and the explosion of mobile devices, the computation demands of data have increased. All compute requests are sent to a remote cloud server for processing, leading to more delay. The emergence of edge computing (EC) can store and compute data generated by network edge devices directly on that device. Nevertheless, implementing compute-intensive tasks on edge devices with limited storage and compute resources makes task allocation more challenging. The complexity of resource allocation can lead to problems such as system performance degradation and time delay. This paper proposes a resource allocation strategy optimized by multi-layer particle swarm optimization (MLPSO) for edge computing to reduce delay. First, we adopt an EC scenario where multiple mobile users are allocated to multiple edge servers dynamically. Then, to construct the delay minimization model, a non-orthogonal multiple access (NOMA) protocol and resource allocation strategy are adopted, which considers the interference between channels. Finally, the delay minimization model is optimized by using MLPSO, that have great potential to help particle swarms jump out of local optima, enhancing the ability to obtain optimal resource allocation strategy for EC. The simulation results, along with comparisons to other methods, clearly demonstrate the effectiveness of the proposed approach in significantly reducing the delay of the EC model. © 2023 IEEE.

关键词： Resource allocation

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：