检索结果-内蒙古大学图书馆

Computer Information Science and Application Technology (CISAT), International Conference on

作者： Zepeng Qiu Lasheng Zhao Ling Wang Tingting Zhang Key Laboratory of Advanced Design and Intelligent Computing Ministry of Education Dalian University Dalian China

ISBN: (数字)9798350375107

ISBN: (纸本)9798350375114

Currently, convolution-based models for keyword spotting focus predominantly on research conducted in clean speech environments. However, their recognition accuracy is lower in low signal-to-noise ratio conditions. To enhance model robustness, we propose the long-term feature fusion module. By extracting and fusing long-term features in both the time and frequency domains, the model’s perception of long-term features is strengthened, improving its recognition performance in noisy environments. At the same time, we propose a band-weighted normalization method, allowing the model to adjust the importance of different frequency bands during the normalization process, further enhancing the robustness of the model. Experimental results indicate that, on the Google Speech Command V2 dataset, the proposed model achieves higher recognition accuracy under various signal-to-noise ratios compared to the comparative models, with a lower parameter count. Moreover, the proposed model exhibits superior generalization to signal-to-noise ratios not covered during the training phase.

关键词： Training Time-frequency analysis Accuracy Convolution Speech recognition Feature extraction Robustness Internet Noise measurement Signal to noise ratio

来源：评论

学校读者我要写书评

暂无评论

Keyword mask pre-training model for the review of standard declaration 4

Keyword mask pre-training model for the review of standard d...

引用

4th International Conference on Computer Engineering and Application, ICCEA 2023

作者： Feng, Chuang Che, Chao Zhang, Qiang Ministry of Education Dalian University Key Laboratory of Advanced Design and Intelligent Computing Dalian116622 China School of Computer Science and Technology Dalian University of Technology Dalian116081 China

ISBN: (纸本)9798350347548

The review of standard declaration of import and export commodities refers to the evaluation of whether the declaration information of the commodities filled in by enterprises conforms to the customs filling standards according to the standardized declaration catalogue The manual review method cannot meet the actual work needs. Therefore, using text classification algorithm to conduct this task is necessary. The classification method based on the pre-training model would rely on the keywords. However, for the review of standard declaration task, excessive reliance on keywords will reduce the classification effect because of ignoring the overall information of the text. Keyword mask pre-training model is proposed. In the fine-tuning stage, the model designs a keyword masking strategy and a keyword blinding strategy. The keyword masking strategy regularizes the BERT model with a mask mechanism, that is, the keywords are reconstructed with other text information except keywords, so that the model focuses on non-keywords information;Keyword blinding strategy makes the model unable to distinguish the categories of text information without contextual text information, which can eliminate the influence of keywords on model decision-making. The experimental results on the real customs declaration data set showing that the method in this thesis exceeds the baseline method in terms of precision and true negative rate. © 2023 IEEE.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

designing Uncorrelated Address Constrain for DNA Storage by DMVO Algorithm

引用

IEEE/ACM Transactions on Computational Biology and Bioinformatics 2022年第2期19卷 866-877页

作者： Cao, Ben Ii, Xue Zhang, Xiaokang Wang, Bin Zhang, Qiang Wei, Xiaopeng Key Laboratory of Advanced Design and Intelligent Computing Ministry of Education Ministry of Education School of Software Engineering Dalian University Dalian116622 China School of Computer Science and Technology Dalian University of Technology Dalian116024 China

At present, huge amounts of data are being produced every second, a situation that will gradually overwhelm current storage technology. DNA is a storage medium that features high storage density and long-term stability and is now considered to be a feasible storage solution. Errors are easily made during the sequencing and synthesis of DNA, however. In order to reduce the error rate, novel uncorrelated address constrain are reported, and a Damping Multi-Verse Optimizer (DMVO)algorithm is proposed to construct a set of DNA coding, which is used as the non-payload. The DMVO algorithm exchanges objects through black/white holes in order to achieve a stable state and adds damping factors as disturbances. Compared with previous work, the coding set obtained by the DMVO algorithm is larger in size and of higher quality. The results of this study reveal that the size of the DNA storage coding set obtained by the DMVO algorithm increased by 4-16 percent, and the variance of the melting temperature decreased by 3-18 percent. © 2004-2012 IEEE.

关键词： Damping

来源：评论

学校读者我要写书评

暂无评论

SC-LSTM Based Human Action Recognition

SC-LSTM Based Human Action Recognition

引用

IEEE International Conference on Signal and Image Processing (ICSIP)

作者： Naixin Kang Pengfei Yi Qingshu Geng Jing Dong Rui Liu Ling Wang Key Laboratory of Advanced Design and Intelligent Computing (Ministry of Education) School of Software Engineering Dalian University Dalian China

ISBN: (数字)9798350350920

ISBN: (纸本)9798350350937

Human Action Recognition is one of the most applied research directions in the field of Computer Vision, which is widely used in human-computer interaction, Augmented Reality (AR) technology, security monitoring, and other scenarios. However, due to the complexity of human action gestures, existing Human Action Recognition methods have certain deficiencies in dealing with variable human gestures and action information, and the accuracy needs to be improved. To improve the accuracy, We propose a multi-dimensional network model based on SC-LSTM(Skip-Connection + LSTM). First, a Temporal Feature Extraction Module is designed based on SC-LSTM, and a Spatial Feature Extraction Module is designed based on CNN and Multi-Attention Mechanism to extract potential human action features from both temporal and spatial dimensions, respectively. Then, a separate SC-LSTM classification network is utilized to process these spatio-temporal features to obtain the final HAR results. The experimental results show that compared to other algorithms, the present model can more fully utilize the information in the temporal dimension, and thus performs better in terms of HAR accuracy.

关键词： Human computer interaction Computer vision Accuracy Image recognition Feature extraction Complexity theory Human activity recognition

来源：评论

学校读者我要写书评

暂无评论

An Active Visual Perception Methodology for the Safety of Human-Robot Collaborations

An Active Visual Perception Methodology for the Safety of Hu...

引用

IEEE International Conference on Signal and Image Processing (ICSIP)

作者： Siyuan Cheng Pengfei Yi Mengmeng Chen Jing Dong Rui Liu Ling Wang Key Laboratory of Advanced Design and Intelligent Computing (Ministry of Education) School of Software Engineering Dalian University Dalian China

ISBN: (数字)9798350350920

ISBN: (纸本)9798350350937

In industrial settings with multiple robotic arms, ensuring human-robot interaction safety through visual feedback is essential. However, existing methods use multiple pre-set cameras to avoid occlusion, making implementation challenging in compact spaces. We propose an adaptive active visual perception method for safety in human-robot collaboration by equipping a single RGBD camera on an idle robotic arm. This method eliminates the need to add additional brackets, is low cost, and is suitable for compact environments compared to current methods. Firstly, to capture comprehensive and accurate global information as much as possible, an active visual perception strategy integrating static visual perception and proprioceptive pose optimization is designed; then, an optimization-based robot position and pose decision method is proposed by using the real-time robots state and obstacles state, which enables the camera at the end of the robot to acquire more effective views; finally, a dual-robot following strategy is proposed to ensure that it can work in tandem. The validity was proved by experimentation. The video are available at https://***/03lwb.

关键词： Visualization Service robots Robot vision systems Human-robot interaction Streaming media Cameras Manipulators

来源：评论

学校读者我要写书评

暂无评论

Acoustic Word Embedding Model with Transformer Encoder and Multivariate Joint Loss

Acoustic Word Embedding Model with Transformer Encoder and M...

引用

intelligent computing and Wireless Optical Communications (ICWOC), International Conference on

作者： Yunyun Gao Qiang Zhang Lasheng Zhao Key Laboratory of Advanced Design and Intelligent Computing Ministry of Education Dalian University Dalian China

As a representation of speech information, acoustic word embedding can enable query-by-example keyword search with low-resource speech data. An acoustic word embedding model with Transformer encoder and multivariate joint loss is presented in this article. First of all, on the basis of BLSTM, the model can extract richer information by adding a Transformer encoder, which improves representation ability of the embedding vector. Then, the contrast loss is improved according to the multiple negative samples, and combined with the hinge loss with distance characteristics, the model generates a more compact vector representation and improves the identification of features. Finally, considering the unbalanced characteristics of samples, the anti-focal loss is introduced to form a multivariate joint loss together with the above two losses, so as to make the model performance better. The model has achieved certain results in word recognition tasks and has certain competitiveness compared with other acoustic word models.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Humanoid Robot Collaborative Lifting Integrating Executable Judgment

Humanoid Robot Collaborative Lifting Integrating Executable ...

引用

2022 IEEE International Symposium on Product Compliance Engineering - Asia, ISPCE-ASIA 2022

作者： Chang, Hua Yi, Pengfei Liu, Rui Dong, Jing Hou, Yaqing Zhou, Dongsheng The Key Laboratory of Advanced Design and Intelligent Computing Ministry of Education School of Software Engineering Dalian University Dalian China Dalian University of Technology Dalian China

ISBN: (数字)9798350332483

ISBN: (纸本)9798350332483

Humanoid robot collaborative lifting can be used in a variety of scenarios that require repetitive lifting tasks. Most existing studies of humanoid robot collaboration often assume that objects can always be lifted, which may result in damage to both the robot and the object if objects are too heavy to lift. To avoid such situations as much as possible, a collaborative lifting approach integrating executable judgment is proposed. First, a target search and localization method is constructed using monocular vision and marker points to identify the task object. Then, an executable judgment strategy is designed to determine whether the object is overweight or not according to robot force analysis. Finally, a multi-robot joint control model is proposed based on collaborative communication to perform collaborative tasks with different loads based on the judgment results. Experiments on two humanoid robots for different types and weights of targets show the effectiveness of the proposed approach. © 2022 IEEE.

关键词： Anthropomorphic robots

来源：评论

学校读者我要写书评

暂无评论

Performance Verification of Smith Predictor Control Using IMC Scheme via Chemical Reaction Networks and DNA Strand Displacement Reaction 9

Performance Verification of Smith Predictor Control Using IM...

引用

9th IEEE Smart World Congress, SWC 2023

作者： Yao, Jingwang Lv, Hui Xiao, Yijun Dalian University Key Laboratory of Advanced Design and Intelligent Computing Ministry of Education School of Software Engineering Dalian116622 China Northeastern University State Key Laboratory of Synthetical Automation for Process Industries Shenyang110004 China

ISBN: (纸本)9798350319804

Biochemical process control is often achieved through feedback control strategies for theoretical analysis and optimization of biochemical reaction processes. Here, the design improves the robustness of the biochemical process by taking advantage of the Smith prediction structure and internal mode control within the chemical reaction network theory framework to intervene against the time delays and perturbations present in the biochemical process. First, an IMC representation of Smith prediction is constructed using the presence of parameter rectification of the control method, and then an IMC-MSP control scheme is proposed. A first-order time-lag controlled object is established in the construction of the IMC-MSP control scheme. Besides, CRNs and kinetic representations are established based on the control structure to facilitate the evaluation of the control system performance. In this paper, CRNs representations of parabolic functions as signal inputs are defined for the first time and optimized for stepping signals to facilitate the design and study of biomolecular circuits. Notably, for the CRNs-based IMC-MSP control scheme, on the one hand, numerical simulation experiments are designed to verify the effectiveness of the whole scheme from multiple perspectives of time delays, external perturbations, and filter tuning parameters;on the other hand, using DNA strand displacement is reflected to verify the feasibility of the whole scheme. © 2023 IEEE.

关键词： Chemical reaction networks (CRNs) DNA strand displacement (DSD) Internal model control (IMC control) Smith predictor

来源：评论

学校读者我要写书评

暂无评论

Learning a Coordinated Network for Detail-Refinement Multiexposure Image Fusion

引用

IEEE Transactions on Circuits and Systems for Video Technology 2023年第2期33卷 713-727页

作者： Li, Jiawei Liu, Jinyuan Zhou, Shihua Zhang, Qiang Kasabov, Nikola K. Ministry of Education School of Software Engineering Dalian University Key Laboratory of Advanced Design and Intelligent Computing Dalian116622 China Dalian University of Technology School of Software Technology Dalian116024 China Auckland University of Technology Knowledge Engineering and Discovery Research Institute Auckland1010 New Zealand Ulster University Intelligent Systems Research Center LondonderryBT52 1SA United Kingdom

Nowadays, deep learning has made rapid progress in the field of multi-exposure image fusion. However, it is still challenging to extract available features while retaining texture details and color. To address this difficult issue, in this paper, we propose a coordinated learning network for detail-refinement in an end-to-end manner. Firstly, we obtain shallow feature maps from extreme over/under-exposed source images by a collaborative extraction module. Secondly, smooth attention weight maps are generated under the guidance of a self-attention module, which can draw a global connection to correlate patches in different locations. With the cooperation of the two aforementioned used modules, our proposed network can obtain a coarse fused image. Moreover, by assisting with an edge revision module, edge details of fused results are refined and noise is suppressed effectively. We conduct subjective qualitative and objective quantitative comparisons between the proposed method and twelve state-of-the-art methods on two available public datasets, respectively. The results show that our fused images significantly outperform others in visual effects and evaluation metrics. In addition, we also perform ablation experiments to verify the function and effectiveness of each module in our proposed method. The source code can be achieved at https://***/lok-18/LCNDR. © 1991-2012 IEEE.

关键词： Extraction

来源：评论

学校读者我要写书评

暂无评论

MSTF: enhancing long-term forecasting with multi-scale temporal fusion in time series forecasting

引用

The Journal of Supercomputing 2025年第9期81卷 1-21页

作者： Zhou, Chengjie Jiang, Ke Liu, Yunjiong Che, Chao Zhang, Qiang School of Computer Science and Technology Dalian University of Technology Dalian China Key Laboratory of Advanced Design and Intelligent Computing Ministry of Education School of Software Engineering Dalian University Dalian China

Long-term time series forecasting plays a vital role in applications like financial market prediction, energy usage forecasting, and traffic flow analysis. Deep learning models, though widely used in this task, often struggle to capture long-term dependencies and periodic patterns due to the overlapping of variables with different periods. To this end, we propose a multi-scale temporal fusion model named MSTF, which extracts sequential features and periodic patterns across multiple scales using a Time Reverse and Transform block and a Dynamic Combination Reconstruction block. Unlike traditional Transformer-based models, MSTF emphasizes overall temporal continuity rather than individual time point values. Experiments on seven datasets against ten advanced models demonstrate MSTF’s superior performance, faster training time, and reduced model complexity, particularly excelling on high-dimensional datasets.

关键词：

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：