检索结果-内蒙古大学图书馆

14th International Symposium on Chinese Spoken Language processing, ISCSLP 2024

作者： Liu, Shuanghong Song, Zhida Fang, Zhihua He, Liang School of Computer Science and Technology Xinjiang University Urumqi China School of Intelligence Science and Technology Xinjiang University Urumqi China Xinjiang Key Laboratory of Signal Detection and Processing Urumqi China Department of Electronic Engineering Beijing National Research Center for Information Science and Technology Tsinghua University Beijing China

ISBN: (纸本)9798331516826

Due to its superior performance and fewer parameters, CAM++ has become the state-of-the-art model for speaker verification tasks. This model uses 2D convolutional blocks to extract front-end features, which are then fed into a densely connected time-delay neural network backbone to extract deep features. However, the simple stacking of 2D convolutions may lead to the generation of a significant amount of redundant features, which is detrimental to efficient feature extraction. Furthermore, although CAM++ already has a relatively small number of parameters, there is still room for further optimization. To address these issues, this paper first employs depthwise separable convolutions to replace the dilated convolutions in the backend network of CAM++, making the model more lightweight. Next, we introduce spatial and channel reconstruction convolution (SCConv) in the ResBlock module of CAM++ to reduce redundant features and optimize the feature extraction process. Finally, after SCConv, we apply squeeze and excitation attention mechanism to model the interdependencies between channels and recalibrate each channel, further enhancing the model’s representational capacity. We name the resulting model LE-CAM++. Our proposed model achieves an EER of 0.686 and a minDCF of 0.084 on the VoxCeleb1-O dataset. Compared to the baseline model CAM++, the EER is reduced by 11%, and the minDCF is reduced by 28%. Additionally, the model parameters are reduced by 8%. ©2024 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Basketball Video Analysis for Automated Game Data Acquisition Deep Learning

Basketball Video Analysis for Automated Game Data Acquisitio...

引用

IEEE International Conference on Electro-information Technology

作者： Diego Rodriguez Garcia Xinrui Yu Jafar Saniie Department of Electrical and Computer Engineering Embedded Computing and Signal Processing (ECASP) Research Laboratory Illinois Institute of Technology Chicago IL U.S.A

ISBN: (数字)9798350330649

ISBN: (纸本)9798350330656

The utilization of data analytics to gain insights into the game of basketball has seen a remarkable surge in the past decade. Leagues such as the National Basketball Association are continuously exploring innovative methods to analyze game data, an approach that has significantly influenced the dynamics of the game. But to perform these analyses, a growing amount of data is needed, which is traditionally annotated by humans. This work proposes a 3-stage system able to automatically acquire relevant basketball game data from a broadcast video. The first stage is an object detector combined with a tracking algorithm to extract the main elements present in a basketball game video. Then, the players' visual information is analyzed to identify the players based on pixel color analysis and number recognition. Finally, a statistics generation algorithm assigns the game events to the corresponding player and team, so that the system can be used as an aid for box score annotation in major leagues, low-cost annotation in amateur games, or in-depth game video analysis.

关键词： Visualization Annotations Image color analysis Data acquisition Optical character recognition Games Detectors

来源：评论

学校读者我要写书评

暂无评论

Advanced Robotic Surveillance for Urban Air Quality Safety

Advanced Robotic Surveillance for Urban Air Quality Safety

引用

IEEE International Conference on Electro-information Technology

作者： Guillaume Hansen James R. Benson Muhammad Rafay Danish Khoa Truong Xinrui Yu Jafar Saniie Department of Electrical and Computer Engineering Embedded Computing and Signal Processing (ECASP) Research Laboratory Illinois Institute of Technology Chicago IL U.S.A.

ISBN: (数字)9798350330649

ISBN: (纸本)9798350330656

This paper presents a multifunctional quadruped robot, specifically engineered for comprehensive air quality monitoring and emergency assistance. Designed to navigate through urban environments, the robot autonomously collects and transmits real-time air quality data to a centralized database. This interdisciplinary paper integrates robotics, environmental science, and emergency response protocols to propose novel solutions for smart city infrastructure, public health, and disaster management. Specifically, this research highlights the critical role that autonomous systems play in monitoring environmental conditions and advancing safety protocols. This research leverages a comprehensive array of sensors including CO2 (MH-Z19), particle (SDS 011), GPS (ADA 746), DHT 22, MQ 9, and ADS 1115 converter, each meticulously implemented to ensure precise environmental data collection and analysis. Furthermore, the integration of cutting-edge technologies such as Robot Operating System (ROS), .NET MAUI, Python, and MariaDB establishes a robust framework for seamless operation, data processing, and secure storage within the quadruped robot system.

关键词： Temperature sensors Protocols Robot sensing systems Dynamic scheduling Sensors Quadrupedal robots Servers

来源：评论

学校读者我要写书评

暂无评论

Autonomous Patrol and Threat Detection Through Integrated Mapping and computer Vision

Autonomous Patrol and Threat Detection Through Integrated Ma...

引用

IEEE International Conference on Electro-information Technology

作者： Robert Soler Alae Moudni Gabriel Roskowski Xinrui Yu Mikhail Gormov Jafar Saniie Department of Electrical and Computer Engineering Embedded Computing and Signal Processing (ECASP) Research Laboratory Illinois Institute of Technology Chicago IL U.S.A.

ISBN: (数字)9798350330649

ISBN: (纸本)9798350330656

This paper presents the creation of an innovative autonomous security robot designed to perform security functions with efficiency and reliability. The robot boasts mapping capabilities, which it utilizes to facilitate autonomous patrol in designated areas. Its primary operations involve the use of computer vision to detect violence, identify weapons and dangerous items, and recognize individuals. Critical incidents are met with an immediate alarm and the subsequent transmission of data to a central security server, which then generates comprehensive reports displayed through a web application for security personnel. The application itself features remote control of the robot, incident report management, status updates, and incident analytics. The robot demonstrates substantial real-world application potential, particularly in crowded environments where it could outperform conventional surveillance. The project combines concepts of engineering, computer science, and cybersecurity, functioning per design but with considerable potential for future refinement and expansion, embodying the concept of an evolving technological solution.

关键词： computer science computer vision Technological innovation Surveillance Weapons Reliability engineering Threat assessment

来源：评论

学校读者我要写书评

暂无评论

Tissue P Systems with Look-Ahead Mode

引用

Chinese Journal of Electronics 2025年第1期23卷 81-86页

作者： Yun Jiang Tao Song Zheng Zhang School of Computer Science and Information Engineering Chongqing Technology and Business University Chongqing China Department of Control Science and Engineering Image Processing and Intelligent Control Key Laboratory of Education Ministry of China Huazhong University of Science and Technology Wuhan China

Tissue P systems are a class of distributed and parallel computing models inspired from inter-cellular communication and cooperation between cells. In this work, a variant of tissue P system, named tissue P system with look-ahead mode, is discussed for decreasing the inherent non-determinism of tissue P systems and helping implementing tissue P systems on computers. Such systems are proved to be universal by simulating register machine, and they are also proved to be able to efficiently solve computationally hard problems by means of a space-time tradeoff, which is illustrated with a polynomial solution to 3-coloring problem.

关键词： computers Computational modeling Medical services Parallel processing Polynomials Computational efficiency Registers Business

来源：评论

学校读者我要写书评

暂无评论

基于语音识别的电磁调控智能超表面

引用

engineering 2023年第3期22卷 185-190页

作者：柏林刘元可徐亮张政王强蒋卫祥仇成伟崔铁军 State Key Laboratory of Millimeter Waves School of Information Science and EngineeringSoutheast UniversityNanjing 210096China Purple Mountain Laboratories Nanjing 211111China Department of Electrical and Computer Engineering National University of SingaporeSingapore 117583Singapore

本文提出并实现了一种基于人类语音识别的智能超表面,用于对电磁波束进行可编程调控。该智能超表面平台由数字编码超表面、语音识别模块、单片机和数模转换器(DAC)电路组成,可根据预先存储的语音指令对电磁波进行智能控制。所构建的数... 详细信息

本文提出并实现了一种基于人类语音识别的智能超表面,用于对电磁波束进行可编程调控。该智能超表面平台由数字编码超表面、语音识别模块、单片机和数模转换器(DAC)电路组成,可根据预先存储的语音指令对电磁波进行智能控制。所构建的数字编码超表面包含6×6个超级子单元,每个超级子单元由4×4个嵌入了变容二极管的有源数字单元组成。语音识别模块配合DAC和单片机对语音指令进行识别,并生成对应的电压序列来控制超表面。此外,在超表面的设计过程中引入遗传算法,可有效优化超表面相位分布。为了验证智能超表面平台的性能,演示了雷达散射截面积缩减、涡旋波束生成和波束分裂三种典型功能。所提出的方案为调控电磁波提供了一种新的途径,并在电磁和声学通信之间架起了连接桥梁。

关键词： Speech recognition Programmable metasurface Genetic algorithm Smart electromagnetic manipulation

来源：评论

学校读者我要写书评

暂无评论

Facial Recognition Attendance Tracking: An Intelligent Monitoring Approach

Facial Recognition Attendance Tracking: An Intelligent Monit...

引用

IEEE International Conference on Electro-information Technology

作者： Syed Tauseeq Hussain Mario Antonio Nogueira Mansur Carvalho Vikrant Kishor Rathod Srinivasa Mani Sankar Reddy Karri Xinrui Yu Mikhail Gromov Jafar Saniie Department of Electrical and Computer Engineering Embedded Computing and Signal Processing (ECASP) Research Laboratory Illinois Institute of Technology Chicago IL U.S.A.

ISBN: (数字)9798350330649

ISBN: (纸本)9798350330656

This paper presents an Intelligent Monitoring System that utilizes the Internet of Things (IoT) and Artificial Intelligence (AI) technologies to automate the class attendance process reliably and efficiently. Conventional approaches for attendance tracking have been laborious and have consumed a significant amount of time, with errors and inconsistencies being a common occurrence. In contrast, the Intelligent Monitoring System combines object detection & recognition AI models, wireless communication, and cloud monitoring to generate reliable attendance data that can be used for various purposes, such as tracking student-by-student attendance data and monitoring overall attendance statistics. The system comprises an ID reader that uses radio frequency tags, a facial recognition system that uses a camera and AI algorithms, and a cloud monitoring system for attendance statistics. The proposed system is designed to overcome the challenges of traditional attendance-taking processes and provide a solution that is accurate, reliable, and efficient.

关键词： Wireless communication Cloud computing Face recognition Scalability Authentication Reliability engineering Security

来源：评论

学校读者我要写书评

暂无评论

Learning Cross-Modal Audiovisual Representations with Ladder Networks for Emotion Recognition

Learning Cross-Modal Audiovisual Representations with Ladder...

引用

International Conference on Acoustics, Speech, and Signal processing (ICASSP)

作者： Lucas Goncalves Carlos Busso Department of Electrical and Computer Engineering Multimodal Signal Processing (MSP) Laboratory The University of Texas at Dallas Richardson TX USA

Representation learning is a challenging, but essential task in audiovisual learning. A key challenge is to generate strong cross-modal representations while still capturing discriminative information contained in unimodal features. Properly capturing this information is important to increase accuracy and robustness in audiovisual tasks. Focusing on emotion recognition, this study proposes novel cross-modal ladder networks to capture modality-specific information while building strong cross-modal representations. Our method utilizes representations from a backbone network to implement unsupervised auxiliary tasks to reconstruct intermediate layer representations across the acoustic and visual networks. The skip connections between the cross-modal encoder and decoder provide powerful modality-specific and multimodal representations for emotion recognition. Our model on the CREMA-D corpus achieves high performance with precision, recall, and F1 scores over 80% on a six-class problem.

关键词： Representation learning Emotion recognition Visualization Focusing Speech recognition Signal processing Acoustics

来源：评论

学校读者我要写书评

暂无评论

Time reversal and Inverse filter for Blind Deconvolution of Ultrasonic Communication Through Solid Channel

Time reversal and Inverse filter for Blind Deconvolution of ...

引用

IEEE International Symposium on Applications of Ferroelectrics (ISAF)

作者： Jafar Saniie Xin Huang Department of Electrical and Computer Engineering Embedded Computing and Signal Processing (ECASP) Research Laboratory Illinois Institute of Technology Chicago IL U.S.A.

ISBN: (数字)9798350371901

ISBN: (纸本)9798350371918

Ultrasonic waves provide an effective means of transmitting information through solid media, such as metal pipes and bars. However, the complex geometry of these materials causes reverberations that degrade the quality of communication. In this paper, we propose denoising approaches using time reversal and inverse filtering methods for blind deconvolution. These methods enable significant improvements in signal-to-noise ratio (SNR), with the inverse filter demonstrating superior performance in both simulated and real-world scenarios. Our results show that this approach provides a robust solution for improving ultrasonic communication in complex solid channels.

关键词： Geometry Deconvolution Filtering Noise reduction Metals Media Solids Reverberation Frequency control Signal to noise ratio

来源：评论

学校读者我要写书评

暂无评论

Dual-Band Diversity Glass Antenna for WiFi Applications

Dual-Band Diversity Glass Antenna for WiFi Applications

引用

2023 Cross Strait Radio Science and Wireless Technology Conference, CSRSWTC 2023

作者： Hu, Peng Fei Leung, Kwok Wa School of Electronics and Information Technology Guangdong Provincial Key Laboratory of Optoelectronic Information Processing Chips and Systems Sun Yat-sen University Guangzhou510006 China City University of Hong Kong State Key Laboratory of Terahertz and Millimeter Waves Department of Electrical Engineering Hong Kong Hong Kong Information and Communication Technology Centre CityU Shenzhen Research Institute Shenzhen518057 China

ISBN: (纸本)9798350358971

A dual-band polarization-diversity omnidirectional glass antenna is presented in this paper. A stepped glass dielectric resonator (DR) is fed by a probe and dual loops for vertical and horizontal polarizations, respectively. This antenna is designed in the 2.4 GHz and 5.8 GHz WiFi bands, with available bandwidths of 7.5% (2.31-2.49 GHz) and 6.5% (5.04-5.38 GHz). Two DR TM modes and two DR TE modes are used to obtain omnidirectional radiation patterns. © 2023 IEEE.

关键词： dielectric resonator antenna Glass antenna omnidirectional antenna

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：