检索结果-内蒙古大学图书馆

Journal of Network Intelligence 2023年第4期8卷 1289-1302页

作者： Huang, Ting-Qing Wang, Chuan-Sheng Chen, Zhao-Qi Zhang, Fu-Quan Meng, Xiang-Long Grau, Antoni Chen, Yang Huang, Jing-Wei College of Computer and Control Engineering Minjiang University Fuzhou University Town No. 200 Xiyuangong Road Fuzhou China Department of Automatic Control Technical Polytechnic University of Catalonia Autonomous Region of Catalonia Barcelona Spain College of Computer and Big Data Fuzhou University Fuzhou University Town No. 2 Wulong Jiangbei Avenue Fuzhou China Digital Media Art Key Laboratory of Sichuan Province Sichuan Conservatory of Music No. 2 Wannianchang Street Chenghua District Sichuan Province Chengdu City China Fuzhou Technology Innovation Center of intelligent Manufacturing information System Minjiang University Fuzhou University Town NO. 200 Xiyuangong Road Fuzhou China Fujian Province University No. 599 Quanxiu Road Juyuanzhou Ecological Civilization District Fujian Province Fuzhou City China College of Electronic Engineering Shandong University of Science and Technology No. 579 Qianwangang Road Huangdao District Qingdao China School of Mechanical and Automotive Engineering Fujian University of Technology No. 33 Xuefu South Road University New District Fujian Province Fuzhou City China

When EEG signals are used to assess the level of student engagement in online teaching tasks, they are often interfered by noise. It is a challenge to effectively remove these noises. Currently, deep learning methods have been applied to the field of EEG denoising. However, existing methods still have some problems. The denoised signals still have obvious noise residue, or the original EEG signals are damaged, and the model fitting speed is too slow. Image dehazing, as a typical denoising task in the field of image enhancement, has achieved great success in recent years. Therefore, inspired by advanced models in this field, we introduce CNN into EEG denoising tasks. In this paper, we take GCANet, an excellent image enhancement model, as an example. The dilated convolutions and gate fusion subnetworks included in GCANet enable more efficient EEG signal denoising. The results demonstrate that the proposed model effectively reduces noise while preserving essential features. Furthermore, in comparison to other state-of-the-art models, our proposed model exhibits enhanced robustness and faster convergence, as evidenced by achieving lower loss values after five epochs. Its good performance provides a new development idea for the field of EEG denoising. © 2023, Taiwan Ubiquitous Information CO LTD. All rights reserved.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Fast extraction of word embedding from Q-contexts

arXiv

引用

arXiv 2021年

作者： Kong, Junsheng Li, Weizhao Liu, Zeyi Liao, Ben Qiu, Jiezhong Hsieh, Chang-Yu Cai, Yi Zhang, Shengyu School of Software Engineering South China University of Technology China University of Cambridge United Kingdom Tencent Quantum Lab China Tsinghua University China The Key Laboratory of Big Data and Intelligent Robot South China University of Technology Ministry of Education

The notion of word embedding plays a fundamental role in natural language processing (NLP). However, pre-training word embedding for very large-scale vocabulary is computationally challenging for most existing methods. In this work, we show that with merely a small fraction of contexts (Q-contexts) which are typical in the whole corpus (and their mutual information with words), one can construct high-quality word embedding with negligible errors. Mutual information between contexts and words can be encoded canonically as a sampling state, thus, Q-contexts can be fast constructed. Furthermore, we present an efficient and effective WEQ method, which is capable of extracting word embedding directly from these typical contexts. In practical scenarios, our algorithm runs 11∼ 13 times faster than well-established methods. By comparing with well-known methods such as matrix factorization, word2vec, GloVe and fasttext, we demonstrate that our method achieves comparable performance on a variety of downstream NLP tasks, and in the meanwhile maintains run-time and resource advantages over all these baselines. © 2021, CC BY.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Noise-Tolerant Learning for Audio-Visual Action Recognition

arXiv

引用

arXiv 2022年

作者： Han, Haochen Zheng, Qinghua Luo, Minnan Miao, Kaiyao Tian, Feng Chen, Yan The National Engineering Lab for Big Data Analytics Xi’an Jiaotong University Xi’an710049 China The School of Computer Science and Technology Xi’an Jiaotong University Xi’an710049 China The Key Laboratory of Intelligent Networks and Network Security Xi’an Jiaotong University Ministry of Education Xi’an710049 China The School of Cyber Science and Engineering Xi’an Jiaotong University Xi’an710049 China

Recently, video recognition is emerging with the help of multi-modal learning, which focuses on integrating distinct modalities to improve the performance or robustness of the model. Although various multi-modal learning methods have been proposed and offer remarkable recognition results, almost all of these methods rely on high-quality manual annotations and assume that modalities among multi-modal data provide semantically relevant information. Unfortunately, the widely used video datasets are usually coarse-annotated or collected from the Internet. Thus, it inevitably contains a portion of noisy labels and noisy correspondence. To address this challenge, we use the audio-visual action recognition task as a proxy and propose a noise-tolerant learning framework to find anti-interference model parameters against both noisy labels and noisy correspondence. Specifically, our method consists of two phases that aim to rectify noise by the inherent correlation between modalities. First, a noise-tolerant contrastive training phase is performed to make the model immune to the possible noisy-labeled data. Despite the benefits brought by contrastive training, it would overfit the noisy correspondence and thus provide false supervision. To alleviate the influence of noisy correspondence, we propose a cross-modal noise estimation component to adjust the consistency between different modalities. As the noisy correspondence existed at the instance level, we further propose a category-level contrastive loss to reduce its interference. Second, in the hybrid-supervised training phase, we calculate the distance metric among features to obtain corrected labels, which are used as complementary supervision to guide the training. Furthermore, due to the lack of suitable datasets, we establish a benchmark of real-world noisy correspondence in audio-visual data by relabeling the Kinetics dataset. Extensive experiments on a wide range of noisy levels demonstrate that our method significant

关键词： Modal analysis

来源：评论

学校读者我要写书评

暂无评论

Progressive Dual Priori Network for Generalized Breast Tumor Segmentation

arXiv

引用

arXiv 2023年

作者： Wang, Li Wang, Lihui Kuai, Zixiang Tang, Lei Ou, Yingfeng Ye, Chen Zhu, Yuemin Wu, Min Shi, Tianliang Engineering Research Center of Text Computing & Cognitive Intelligence Ministry of Education Key Laboratory of Intelligent Medical Image Analysis and Precise Diagnosis of Guizhou Province State Key Laboratory of Public Big Data College of Computer Science and Technology Guizhou University Guiyang550025 China Imaging Center Harbin Medical University Cancer Hospital Harbin150081 China Radiology Department Guizhou Provincial People’s Hospital Guiyang550002 China Radiology Department Tongren People’s Hospital Tongren554300 China Univ Lyon INSA Lyon CNRS Inserm CREATIS UMR 5220 U1206 LyonF-69621 France

To promote the generalization ability of breast tumor segmentation models, as well as to improve the segmentation performance for breast tumors with smaller size, low-contrast and irregular shape, we propose a progressive dual priori network (PDPNet) to segment breast tumors from dynamic enhanced magnetic resonance images (DCE-MRI) acquired at different centers. The PDPNet first cropped tumor regions with a coarse-segmentation based localization module, then the breast tumor mask was progressively refined by using the weak semantic priori and cross-scale correlation prior knowledge. To validate the effectiveness of PDPNet, we compared it with several state-of-the-art methods on multi-center datasets. The results showed that, comparing against the suboptimal method, the DSC and HD95 of PDPNet were improved at least by 5.13% and 7.58% respectively on multi-center test sets. In addition, through ablations, we demonstrated that the proposed localization module can decrease the influence of normal tissues and therefore improve the generalization ability of the model. The weak semantic priors allow focusing on tumor regions to avoid missing small tumors and low-contrast tumors. The cross-scale correlation priors are beneficial for promoting the shape-aware ability for irregular tumors. Thus integrating them in a unified framework improved the multicenter breast tumor segmentation performance. The source code and open data can be accessed at https://github. com/wangli100209/PDPNet. Copyright © 2023, The Authors. All rights reserved.

关键词： Magnetic resonance

来源：评论

学校读者我要写书评

暂无评论

An Adaptive Irregular Convolution U-Net for Reconstructing Ancient Channel of the Yellow River

An Adaptive Irregular Convolution U-Net for Reconstructing A...

引用

IEEE SENSORS

作者： Zhishun Guo Jianhui Zhao Ning Li Lin Wu College of Computer and Information Engineering Henan University Kaifeng China Henan. Key Laboratory of Big Data Analysis and Processing Henan University Kaifeng China Henan Engineering Research Center of Intelligent Technology and Application Henan University Kaifeng China

The lower reaches of the ancient Yellow River (AYR) migrated very frequently, like a loong swinging its tail on the land of China. This paper, based on Gaofen-3 full-polarization synthetic aperture radar (SAR) data, analyzed scattering characteristics of archaeolandscapes, which provided reliable evidences for the existence of AYR between Lankao and Sui County. Then, an adaptive irregular convolution kernel U-Net (AICK-U-Net) was proposed to reconstruct the part of channel of AYR, based on the image obtained by polarization decomposition, and the precision reached 96.21%. In summary, polarization decomposition and neural network in spaceborne SAR field, provided new insights for detecting and reconstructing archaeolandscapes.

关键词： Convolution Spaceborne radar Neural networks Scattering Sensors Rivers Reliability

来源：评论

学校读者我要写书评

暂无评论

Efficient EK-means:Extended K-means Clustering for Categorical data with High Processing Speed

Efficient EK-means:Extended K-means Clustering for Categoric...

引用

作者： Chen Xi Peng Shuo HunanProvincial Key Laboratory of Intelligent Processing of Big Data on Transportation School of computer and communication engineeringChangsha University of Science&technology

The typical representative of the hard clustering algorithm,K-means,is one of the fastest processing algorithms with good ***,it cannot deal with categorical attributes,which is one of the important indicators to measure the pros and *** to the lack of processing capabilities on categorical attributes,k-means has a large limit on data processing *** paper proposes a clustering algorithm extends from *** algorithm introduces the concept of a Pseudo-mean distance calculation formula and a counting-table so that categorical attributes can be processed while reducing the time cost as much as *** results illustrate the proposed Pseudo-means can extend the processing range of k-means to category-type data,and the counting table also effectively reduces the time cost.

关键词： Cluster Efficient EK-means:Extended K-means Clustering for Categorical data with High Processing Speed

来源：评论

学校读者我要写书评

暂无评论

Exploration of Novel Nsm Methods to Achieve Stronger Frps with Minimal Epoxy

SSRN

引用

SSRN 2024年

作者： Yan, Handong Yin, Jianli Zhu, Xinan Lou, Tiejiong Li, Haifeng Sun, Wei College of Civil Engineering Huaqiao University Xiamen361021 China Higher-educational Engineering Research Centre for Intelligence and Automation in Construction of Fujian Province College of Civil Engineering Huaqiao University Xiamen361021 China Key Laboratory for Intelligent Infrastructure and Monitoring of Fujian Province College of Civil Engineering Huaqiao University Xiamen361021 China KZJ New Materials Group Co. Ltd. Xiamen361101 China Shandong Province Big Data Center Shandong Jinan250000 China University of Coimbra Department of Civil Engineering Cemmpre Arise 3030-788 Portugal

Conventional NSM methods require a significant amount of epoxy to achieve the desired bond strength, leading to substantial material costs and carbon emissions. Anchorages offer an opportunity to fully develop the tensile strength of NSM FRPs and reduce epoxy consumption. However, there is a lack of comprehensive studies to guide the application of this promising technology. Unlike conventional pull-off tests, a series of bending experiments were carried out to investigate the direct impacts of groove arrangement, bond condition, anchorage, and FRP cross-sectional area on the failure modes and load capacities of flexural elements. The findings were then utilized to calibrate numerical models, which could be directly integrated into equilibrium equations to predict the load-deflection responses for the concrete elements strengthened with NSM methods. FRP models were developed to capture the softening behavior caused by slip. Cracking models that incorporated the tensile contributions of filler material and cracked concrete was proposed to describe the post-cracking behavior. Failure modes such as FRP debonding and rupture related to specific FRP arrangements were employed to determine the ultimate load capacities. The outcomes indicate that anchored NSM methods improve FRP utilization and flexural strength in comparison to traditional NSM techniques, while also reducing epoxy consumption. © 2024, The Authors. All rights reserved.

关键词： Numerical models

来源：评论

学校读者我要写书评

暂无评论

Weakly-supervised multi-granularity map learning for vision-and-language navigation 22

Weakly-supervised multi-granularity map learning for vision-...

引用

Proceedings of the 36th International Conference on Neural Information Processing Systems

作者： Peihao Chen Dongyu Ji Kunyang Lin Runhao Zeng Thomas H. Li Mingkui Tan Chuang Gan South China University of Technology and Information Technology R&D Innovation Center of Peking University South China University of Technology South China University of Technology and Pazhou Laboratory Shenzhen University Information Technology R&D Innovation Center of Peking University South China University of Technology and Key Laboratory of Big Data and Intelligent Robot Ministry of Education MIT-IBM Watson AI Lab and UMass Amherst

ISBN: (纸本)9781713871088

We address a practical yet challenging problem of training robot agents to navigate in an environment following a path described by some language instructions. The instructions often contain descriptions of objects in the environment. To achieve accurate and efficient navigation, it is critical to build a map that accurately represents both spatial location and the semantic information of the environment objects. However, enabling a robot to build a map that well represents the environment is extremely challenging as the environment often involves diverse objects with various attributes. In this paper, we propose a multi-granularity map, which contains both object fine-grained details (e.g., color, texture) and semantic classes, to represent objects more comprehensively. Moreover, we propose a weakly-supervised auxiliary task, which requires the agent to localize instruction-relevant objects on the map. Through this task, the agent not only learns to localize the instruction-relevant objects for navigation but also is encouraged to learn a better map representation that reveals object information. We then feed the learned map and instruction to a waypoint predictor to determine the next navigation goal. Experimental results show our method outperforms the state-of-the-art by 4.0% and 4.6% w.r.t. success rate both in seen and unseen environments, respectively on VLN-CE dataset.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Sequence-to-Sequence Text Summarization Model with Topic Based Attention Mechanism 16th

A Sequence-to-Sequence Text Summarization Model with Topic B...

引用

16th Web Information Systems and Applications Conference, WISA 2019

作者： Pan, Heng-Xi Liu, Hai Tang, Yong South China Normal University Guangzhou510631 China Guangzhou Key Laboratory of Big Data Intelligent in Education Guangzhou510631 China

ISBN: (纸本)9783030309510

One of the limitation of automatic summarization is that how to take into account and reflect the implicit information conveyed between different text and the scene influence. In particularly, the generation of news headlines should under specific scene and topic in the field of journalism. Traditionally, Sequence-to-Sequence (Seq2Seq) with attention model has shown great success in summarization. However, Sequence-to-Sequence (Seq2Seq) with attention model is focusing on features of the text only, not on implicit information between different text and scene influence. In this work, we present a combination of techniques that harness scene information which reflects by word topic distribution to improve abstractive sentence summarization. This model combines word topic distribution of LDA topic model as an external attention mechanism to better text summarization result. This model contains an RNN network as an encoder and decoder part, encoder is used to embed original text in a low dimensional dense vector as previous works, and decoder uses attention mechanism to incorporate word-topic distribution and low dimensional dense vector of encoder. The proposed approach is evaluated by datasets of CNN/Daily Mail and citation. The result shows that it is better than the aforementioned methods. © 2019, Springer Nature Switzerland AG.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

Memory-efficient hierarchical neural architecture search for image restoration

arXiv

引用

arXiv 2020年

作者： Zhang, Haokui Li, Ying Chen, Hao Gong, Chengrong Bai, Zongwen Shen, Chunhua Northwestern Polytechnical University China Intellifusion China Shaanxi Key Laboratory of Intelligent Processing for Big Energy Data Yan'an University China

Recently, much attention has been spent on neural architecture search (NAS), aiming to outperform those manually-designed neural architectures on high-level vision recognition tasks. Inspired by the success, here we attempt to leverage NAS techniques to automatically design efficient network architectures for low-level image restoration tasks. In particular, we propose a memory-efficient hierarchical NAS (termed HiNAS) and apply it to two such tasks: image denoising and image super-resolution. HiNAS adopts gradient based search strategies and builds a flexible hierarchical search space, including the inner search space and outer search space. They are in charge of designing cell architectures and deciding cell widths, respectively. For the inner search space, we propose a layer-wise architecture sharing strategy (LWAS), resulting in more flexible architectures and better performance. For the outer search space, we design a cell-sharing strategy to save memory, and considerably accelerate the search speed. The proposed HiNAS method is both memory and computation efficient. With a single GTX1080Ti GPU, it takes only about 1 hour for searching for denoising network on the BSD-500 dataset and 3.5 hours for searching for the super-resolution structure on the DIV2K dataset. Experiments show that the architectures found by HiNAS have fewer parameters and enjoy a faster inference speed, while achieving highly competitive performance compared with state-of-the-art methods. © 2020, CC BY-NC-ND.

关键词： Image reconstruction

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：