检索结果-内蒙古大学图书馆

IEEE International Conference on Information Communication and signal processing (ICICSP)

作者： Lulu Ye Huabin Wang Yufeng Shi Liang Tao Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education Anhui University Hefei China

At present, the biometric template algorithms with state of the art recognition performance cannot guarantee the security of the system, whereas the security biometric template algorithms usually suffer from inaccurate recognition. In order to improve both the recognition performance and security of feature template for finger vein, a cancelable template algorithm is proposed in this paper. Firstly, a variable curvature Gabor filter is designed for finger vein orientation feature extraction. The maximum and sub-maximum orientation of Gabor filter are combined with differential excitation respectively to generate joint distribution features, which serve as feature vectors of finger vein. Secondly, the extracted feature vectors are reduced dimension by PCA, and the reduced feature vectors are combined with the pseudo random matrix, which is produced by a token to generate a cancelable template to guarantee the security. Finally, cancelable templates in two orientations are fused to increase the irreversibility of templates by using the improved canonical correlation analysis. Extensive experiments on the SDUMLA-FV and PolyU databases demonstrate the superiority of the proposed method in terms of verification performance, diversity, revocability/reusability and irreversibility.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Low-Complexity and High-Efficient Pilot Allocation Scheme Based on User Grouping in Massive MIMO System

A Low-Complexity and High-Efficient Pilot Allocation Scheme ...

引用

International Conference on Communication Technology (ICCT)

作者： Xinying Zhao Yijing Yao Lei Xu Hongyu Fang Hongwei Zhang Xiaohui Li Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education Anhui University Hefei China

Pilot allocation is one of the effective means to reduce pilot pollution in massive Multiple-Input Multiple-Output (MIMO) systems. The goal of this paper is to improve the uplink achievable sum rates of strong users, and ensure the quality of service (QoS) requirements of weak users at the same time, so that the sum rates of system can be improved. Combining with the technical advantage of pilot grouping, a low complexity pilot allocation scheme based on matching algorithm is proposed, which divides the users in the target cell into weak user group and strong user group, and adopts the minimum-maximum matching method to allocate pilots in weak user group. Through the introduction of Hungarian algorithm, a pilot allocation method is designed to ensure the fairness of the strong users. The simulation results show that, compared with the smart pilot allocation scheme, the pilot allocation scheme based on Hungarian algorithm, the pilot allocation scheme based on user grouping and the random pilot allocation scheme, the system performance of the proposed scheme has been effectively improved.

关键词： Resource management Pollution Uplink Interference Fading channels Antennas Massive MIMO

来源：评论

学校读者我要写书评

暂无评论

3R: Word and Phoneme Edition based Data Augmentation for Lexical Punctuation Prediction

3R: Word and Phoneme Edition based Data Augmentation for Lex...

引用

International Conference on Computational Intelligence and Security

作者： Aihua Zheng Naipeng Ye Xiao Wang Xiao Song Key Lab of Intelligent Computing and Signal Processing of Ministry of Education Hefei China Peking University Shenzhen Institute Shenzhen China Anhui Provincial Key Laboratory of Multimodal Cognitive Computation School of Computer Science and Technology Anhui University Hefei China

Existing Lexical Punctuation Prediction methods are mainly trained on the standard clean data while losing the generalization in practical automatic speech recognition (ASR) system with ubiquitous transcription errors. To bridge the gap between clean training data and noisy testing data, we propose three random (3R) data augmentation strategies: random word deletion (RWD), random word substitution (RWS), and random phoneme edition (RPE) in both word and phoneme levels on the training dataset. Specifically, we contribute an acoustically similar vocabulary with phoneme level editions for acoustically similar word substitution. In addition, we first introduce the RoBERTa-large model into a punctuation prediction task to capture the semantics and the long-distance dependencies in language. Extensive experiments on the English dataset IWSLT2011 yield to a new state-of-the-art comparing to the prevalent punctuation prediction methods.

关键词： Training Vocabulary Semantics Training data Predictive models Noise measurement Security

来源：评论

学校读者我要写书评

暂无评论

Hybrid graph-based radiology report generation

引用

Expert Systems with Applications 2025年 289卷

作者： Dengdi Sun Chaofan Mu Xuyang Fan Jin Tang Zegeng Li Jie Zhu Zhuanlian Ding Key Laboratory of Intelligent Computing & Signal Processing (ICSP) Ministry of Education School of Artificial Intelligence Anhui University Hefei 230601 China Anhui Provincial Key Laboratory of Multimodal Cognitive Computation School of Computer Science and Technology Anhui University Hefei 230601 China First Affiliated Hospital of Anhui University of Chinese Medicine Hefei 230031 China College of Integrated Chinese and Western Medicine Anhui University of Chinese Medicine Hefei 230012 China School of Internet Anhui University Hefei 230039 China

Automatic report generation plays a crucial role in clinical practice by alleviating the heavy workload on doctors and helping to prevent misdiagnoses or missed diagnoses. In the context of radiology report generation, knowledge injection is essential, particularly with the encoder-decoder framework commonly used in image captioning tasks. However, existing studies predominantly rely on expert knowledge, which is both challenging to collect and lacks universality. Additionally, this expert knowledge typically influences single-mode information and overlooks the importance of bridging the visual-to-textual gap. To address these challenges, we propose a hybrid graph-based approach for radiology report generation. Our method integrates two key components: the semantic homogeneous graph (SHG) and the cross-modal heterogeneous graph (CHG). Specifically, the SHG is constructed by mining semantic relationships between keywords across the entire corpus to generate universal knowledge. The CHG, conversely, is built from visual features, textual features, and corresponding knowledge embeddings, enabling knowledge injection during modal interaction. By leveraging graph convolutional networks to enhance graph embeddings, our model improves the quality of generated reports. Experimental results on two widely used benchmark datasets, IU-Xray and MIMIC-CXR, demonstrate the effectiveness of our approach. Notably, our method achieves a BLEU-4 score of 0.185 on the IU-Xray dataset and an F1 score of 0.418 on the MIMIC-CXR dataset, significantly outperforming existing methods.

关键词： Graph convolutional network Hybrid graph Knowledge injection Radiology report generation

来源：评论

学校读者我要写书评

暂无评论

Correction: Self-orthogonal codes over a non-unital ring and combinatorial matrices

引用

Designs, Codes and Cryptography 2023年第2期91卷 691-691页

作者： Shi, Minjia Wang, Shukai Kim, Jon-Lark Solé, Patrick Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education School of Mathematical Sciences Anhui University Hefei China School of Mathematical Sciences Anhui University Hefei China Department of Mathematics Sogang University Seoul South Korea Aix Marseille Univ CNRS Centrale Marseille I2M Marseilles France

来源：评论

学校读者我要写书评

暂无评论

Seismic Event Detection via Deep Multi-Task Learning

Seismic Event Detection via Deep Multi-Task Learning

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Yang Yu Lei Zhang Jiakai Shen Qingcai Wang Guiquan Liu School of Computer Science and Technology University of Science and Technology of China Hefei China Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education Anhui University Hefei China Department of Computer Science and Technology Nanjing University Nanjing China

ISBN: (数字)9781728169262

ISBN: (纸本)9781728169279

It is fundamental to detect seismic events reliably and efficiently when processing continuous waveform data recorded by seismic stations. Recently, convolutional neural network (CNN) based detecting methods have been proposed for seismic events detection and obtained great success in this area, where the learning of seismic event detecting network of all seismic stations is considered as one learning task and numerous labeled data need to be collected for training the detecting network. However, they tend to ignore the differences between seismic stations caused by geographic position. Moreover, due to a few seismic activities and high cost of manual data labeling, in some areas, the labeled data for seismic event detecting tasks is insufficient. Under this condition, these methods always encounter over-fitting problem leading to bad detection performance. In this paper, we propose a multi-task based framework based on convolutional neural network for accurate seismic event detection under the condition of insufficient labeled data. Specifically, we first cluster the seismic stations into several station clusters and treat the learning of seismic event detecting network of every station cluster as a learning task, and then we propose a deep multi-task network named detectMTIA among multiple tasks. Experimental results on a real-world seismic dataset with nine stations demonstrate the effectiveness of the proposed framework, especially when the labeled data is insufficient.

关键词： Task analysis Clustering algorithms

来源：评论

学校读者我要写书评

暂无评论

Exploring cascaded write amplification in LSM-tree based key-value stores with solid-state disks 21

Exploring cascaded write amplification in LSM-tree based key...

引用

21st IEEE International Conference on High Performance computing and Communications, 17th IEEE International Conference on Smart City and 5th IEEE International Conference on Data Science and Systems, HPCC/SmartCity/DSS 2019

作者： Sun, Hui Dai, Shangshang Huang, Jianzhong Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education School of Computer Sci. and Tech. Anhui University Hefei China Wuhan National Lab. for Optoelectronics Huazhong University of Sci. and Tech Wuhan China

ISBN: (纸本)9781728120584

Log-structured merge tree (i.e., LSM-tree) based key-value stores, which are widely used in big-data applications, provide high performance. NAND Flash-based Solid state disks (i.e., SSDs) become the popular devices that alternative to hard disk drives (i.e., HDDs) because of the high performance and low power consumption. LSM-tree KV stores with SSDs are deployed in large-scale storage systems to achieve high performance in the cloud. However, the host-side write amplification in LSM-tree KV stores and the device-side write amplification in SSDs can impair the overall system performance. The former that is attributed to compaction has a burn on I/O bandwidth between two sides. The latter, which results from garbage collection in NAND Flash memory, blocks user I/Os from the host and degrades the SSD performance. In this study, we explored cascaded write amplification in LSM-tree KV stores with SSDs. Our primary goal is to comprehensively study two-level cascaded write amplification on the host and device sides. We quantitatively analyze the impact of two-level write amplification on overall performance. This work reveals details of cascaded write amplification and helps developers in the software stacks of LSM-tree based KV stores with SSDs. © 2019 IEEE.

关键词： NAND circuits

来源：评论

学校读者我要写书评

暂无评论

LCD codes from tridiagonal Toeplitz matrice

arXiv

引用

arXiv 2021年

作者： Shi, Minjia Özbudak, Ferruh Xu, Li Solé, Patrick Key Laboratory of Intelligent Computing Signal Processing Ministry of Education School of Mathematical Sciences Anhui University HefeiAnhui230601 China Department of Mathematics Institute of Applied Mathematics Middle East Technical University Ankara Turkey School of Mathematical Sciences Anhui University HefeiAnhui230601 China I2M Aix Marseille Univ. Centrale Marseille CNRS Marseille France

Double Toeplitz (DT) codes are codes with a generator matrix of the form (I, T) with T a Toeplitz matrix, that is to say constant on the diagonals parallel to the main. When T is tridiagonal and symmetric we determine its spectrum explicitly by using Dickson polynomials, and deduce from there conditions for the code to be LCD. Using a special concatenation process, we construct optimal or quasi-optimal examples of binary and ternary LCD codes from DT codes over extension *** Codes 94B05, 15B05, 12E10 Copyright © 2021, The Authors. All rights reserved.

关键词： Matrix algebra

来源：评论

学校读者我要写书评

暂无评论

Reversible data hiding in encrypted images based on pixel prediction and bit-plane compression

arXiv

引用

arXiv 2019年

作者： Yin, Zhaoxia Peng, Yinyin Xiang, Youzhi school of Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Anhui University Hefei230601 China

Reversible data hiding in encrypted images (RDHEI) receives growing attention because it protects the content of the original image while the embedded data can be accurately extracted and the original image can be reconstructed lossless. To make full use of the correlation of the adjacent pixels, this paper proposes an RDHEI scheme based on pixel prediction and bitplane compression. Firstly, to vacate room for data embedding, the prediction error of the original image is calculated and used for bit-plane rearrangement and compression. Then, the image after vacating room is encrypted by a stream cipher. Finally, the additional data is embedded in the vacated room by multi-LSB substitution. Experimental results show that the embedding capacity of the proposed method outperforms the state-of-the-art methods. Copyright © 2019, The Authors. All rights reserved.

关键词： Steganography

来源：评论

学校读者我要写书评

暂无评论

Author Correction: Full reconstruction of simplicial complexes from binary contagion and Ising data

引用

Nature communications 2022年第1期13卷 3692页

作者： Huan Wang Chuang Ma Han-Shuang Chen Ying-Cheng Lai Hai-Feng Zhang The Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education School of Mathematical Science Anhui University Hefei 230601 China. School of Internet Anhui University Hefei 230601 China. School of Physics and Material Science Anhui University Hefei 230601 China. School of Electrical Computer and Energy Engineering Arizona State University Tempe AZ 85287 USA. The Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education School of Mathematical Science Anhui University Hefei 230601 China. haifengzhang1978@***.

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：