检索结果-内蒙古大学图书馆

Artificial Intelligence and Electromechanical Automation (AIEA), International Conference on

作者： Shixin Chen Department of Computer Science Software Engineering Institute of Guangzhou Guangzhou China

ISBN: (数字)9798350366174

ISBN: (纸本)9798350366181

Owing to the challenge of target occlusion leading to tracking failure during the target tracking process, achieving efficient and robust tracking of targets under occlusion scenarios has become a focal point of research. In this paper, we addresses the issue of tracking failure caused by the reduction or disappearance of target appearance information in occlusion scenarios. It proposes a study on target tracking under occlusion scenes based on deep learning methodologies. A foundation tracker is constructed that leverages location enhancement and multi-template fusion. Built upon this, graph attention mechanisms and target loss detection mechanisms are introduced. In cases where the target is lost due to occlusion, a re-detection strategy is employed to resume tracking once the target reappears. Experiments conducted on datasets demonstrate that the proposed algorithm can effectively determine whether a target is undergoing occlusion and significantly enhances tracking performance under such conditions.

关键词： Deep learning Target tracking Correlation Automation Attention mechanisms Neural networks filtering algorithms Artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

A Machine Learning Model for Recognition and Classification of Emotion Using Facial Expressions

A Machine Learning Model for Recognition and Classification ...

引用

Delhi Section Conference (DELCON), IEEE

作者： R. Kamalraj Sandeep Kumar Jain Trapty Aggarwal Khushmeen Kaur Brar Mohammad Al-Farouni Anita Gehlot Department of CSE and IT Jain Deemed to be University Bangalore India Department of Electrical Engineering Vivekananda Global University Jaipur India MSET Maharishi University of IT Uttar Pradesh India Department of Electronics and Communication Engineering Chandigarh University Mohali Punjab India The Islamic University Najaf Iraq Department of ECE Uttaranchal Institute of Technology Dehradun India

ISBN: (数字)9798331518592

ISBN: (纸本)9798331518608

Emotion recognition has become a major area of multidisciplinary research in recent years and has many potential applications in the development of artificial intelligence and human-computer interface related design of human health detection. Expansion experiments are carried out by researchers using various stimuli, attributes, and categorization methods. This paper makes use of hybrid feature optimization algorithms i. e, IWSS Embedded and PSO for recognizing emotions using facial expressions. A pretrained model ResNet50V2 is used as a feature extractor and an image filter called Discretization is applied in order to improve the accuracy of the proposed approach. The results revealed that the methodology attained an overall accuracy of 75%.

关键词： Emotion recognition Accuracy Face recognition filtering algorithms Multilayer perceptrons Feature extraction Brain modeling Electroencephalography Optimization Load modeling

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Adaptive Technique for Barker Code Matched Filter Sidelobe Cancellation

A Hybrid Adaptive Technique for Barker Code Matched Filter S...

引用

Telecommunications Conference (ITC-Egypt), International

作者： Hazem Kamel Mohamed G. Shehata Samer Emad Eldin Ali Radar Department Military Technical College Cairo Egypt

ISBN: (数字)9798350351408

ISBN: (纸本)9798350351415

Barker phase coding is a well-known technique utilized in pulse compression radars employing binary phase-coded waveforms. Despite its effectiveness, a common issue associated with this method is the generation of undesired sidelobes in the output of the sub-pulse matched filter. These sidelobes have the potential to obscure the reflected echoes from nearby weak targets, consequently hampering the radar's detection capabilities. To address this challenge, this study introduces an innovative approach that combines the use of an Adaptive Normalized Least Mean Square (NLMS) filter for noise reduction and a Recursive-Least-Square (RLS) filter to mitigate the sidelobes produced by the matched filter. The integration of the RLS filter with the NLMS filter presents a promising strategy for combating radar active jamming. Through extensive simulation testing, the performance of this hybrid filtering technique is assessed in comparison to an optimal filter. The results indicate that both filters exhibit robust performance under high signal-to-noise ratios (SNRs). However, the proposed adaptive filter demonstrates superior performance over the optimal filter, especially in environments characterized by low signal-to-noise ratios (SNRs). This capability is particularly advantageous for detecting targets with low SNR levels, such as small Radar Cross Section (RCS) targets.

关键词： Matched filters Radar cross-sections Codes Noise reduction Adaptive filters Object detection filtering algorithms

来源：评论

学校读者我要写书评

暂无评论

Noise Reduction of Bird Calls Based on a Combination of Spectral Subtraction, Wiener filtering, and Kalman filtering

Noise Reduction of Bird Calls Based on a Combination of Spec...

引用

IEEE International Conference on Automation, Electronics and Electrical Engineering (AUTEEE)

作者： Zenghui Chen College of Photonic and Electronic Engineering Fujian Normal University Fujian China

ISBN: (数字)9781665471978

ISBN: (纸本)9781665471985

Speechis a channel of communication for birds, and by analyzing bird calls, biologists can obtain much important biological information. However, in the speech recognition process, there is inevitably the influence of environmental noise. However, the recognition rate in the presence of ambient noise is not yet satisfactory, so strengthening the noise immunity is a problem that needs further research and solution in the speech recognition process. This paper proposes a bird call noise reduction method based on a combination of spectral subtraction, Wiener filtering, and Kalman filtering to enhance the noisy speech information. In this paper, through white noise addition and using three methods of independent use and combined into one filter use, through the listening side, image analysis, signal-to-noise ratio, and coherence analysis of out, the noise reduction effect is very significant, and can be applied to the background noise filtering of the real environment in practice, for human understanding of bird habits a step further.

关键词： Wiener filters Speech recognition filtering algorithms White noise Birds Information filters Biology

来源：评论

学校读者我要写书评

暂无评论

Graph Convolutional Networks for Designing Collaborative filtering-Based Health Recommender Systems

Graph Convolutional Networks for Designing Collaborative Fil...

引用

International Conference on New Technologies of Information and Communication (NTIC)

作者： Boudjemaa Boudaa Imen Bestani Noureddine Benadjrouda Department of Computer Science University of Tiaret Tiaret Algeria

Recommender systems provide useful item suggestions (products or services) to users as part of their decision-making processes. The effectiveness of recommender systems is now clearly confirmed in various fields of application (e.g., YouTube, Amazon, Facebook, ResearchGate). In the literature, many research works have addressed the application of recommendations in the field of health in what are called health recommender systems (HRS). HRS is an innovative alternative when it comes to providing information to help doctors in the diagnosis/treatment of diseases, as well as helping patients with recommendations on how to maintain their well-being. However, the proposed development approaches in this field are limited to traditional models that lack the accuracy and effectiveness, which are vital in healthcare. This paper presents a design model for collaborative filtering-based health recommender systems using graph neural networks (GNN) via its promising Graph Convolutional Network (GCN) architecture. In this model, the convolution layer works with a simplified and efficient GCN algorithm named LightGCN. GCN-based methods are among the new cutting-edge approaches in recommender systems, and LightGCN has proven its superiority in recommendation accuracy.

关键词： Video on demand Social networking (online) Decision making Collaboration Medical services filtering algorithms Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Distributed State and Fault Fusion filtering for Nonlinear Systems Subject to Quantization Effects

Distributed State and Fault Fusion Filtering for Nonlinear S...

引用

IEEE International Conference on Control Science and Systems Engineering (CCSSE)

作者： Zhibin Hu Jun Hu Junhua Du Yuna Zhao Department of Mathematics Heilongjiang Provincial Key Laboratory of Optimization Control and Intelligent Analysis for Complex Systems Harbin University of Science and Technology Harbin China

In this paper, the distributed state and fault fusion filtering problem is discussed for a class of multi-sensor networked systems (MSNSs) with randomly occurring quantized measurements (ROQMs). A series of random variables obeying the Bernoulli distribution are adopted to depict the phenomenon of the ROQMs and the fault signal is assumed to appear with an unknown constant type. The aim of this paper is to provide the state and fault fusion filtering algorithm by using the inverse covariance intersection fusion method. Moreover, a proper filter gain is designed by minimizing the local upper bound of filtering error covariance. Finally, the effectiveness of the presented fusion method is verified by using an illustrative example.

关键词： Upper bound Quantization (signal) filtering filtering algorithms Systems engineering and theory Numerical simulation Control systems

来源：评论

学校读者我要写书评

暂无评论

Multi Points Updated and Distance Filtered Kriging Surrogate Model: Application in EOSS Optimization

引用

Chinese Journal of Electronics 2023年第1期22卷 209-213页

作者： Xiaolu Liu Yingwu Chen Yingguo Chen Renjie He School of Information System and Management National University of Defense Technology Changsha China

We put forward a multi points updated and distance filtered Kriging surrogate model. When building Kriging models, two update approaches are used to select infilling points: optimal points and maximized expected improvement. We use Improved general pattern search (IGPS) algorithm to get these points. In IGPS, search step is substituted by GA and SQP and poll step is retained. To decrease simulation times, we adopt a distance filter to eliminate potentially replicated samples. A satellite orbit parameter optimization problem is formulated, which is solved by the proposed method and STK/Analyzer respectively. The results showed that Kriging models, which use multi update points and distance filter, yield global approximations that are more accurate than Analyzer.

关键词： Earth Analytical models Satellites Buildings filtering algorithms Orbits Optimization

来源：评论

学校读者我要写书评

暂无评论

Enhanced Time-Optimal Filter-Based Point-to-Point Trajectory Generation with Vibration Suppression

Enhanced Time-Optimal Filter-Based Point-to-Point Trajectory...

引用

Mechatronics and Robotics Engineering (ICMRE), International Conference on

作者： Rajanya Yalamanchili Stefan Schlee Markus Märkl Frank A. King Faculty of Engineering Sciences Technical University of Applied Sciences Rosenheim Rosenheim Germany Chair of High-Power Converter Systems TUM School of Engineering and Design Technical University of Munich Munich Germany

ISBN: (数字)9798350394276

ISBN: (纸本)9798350394283

This paper proposes enhancements to FIR filter-based trajectory generation for time-optimal Point-to-Point movements of machines with residual vibration suppression. Considering a set of kinematic constraints and the relevant eigenfrequencies of the mechanical subsystem, filter parameters for a series of rectangular filters are computed. With the proposed enhancements to the filter merge algorithm for eigenmode suppression, the trajectories generated are always compliant with the kinematic constraints and utilize the full potential of the kinematic limits. It is further extended to include a Zero-Vibration (ZV) input shaper to generate time-optimal trajectories in the cases when filter merge algorithms fail to compute a time-optimal filter parameter set. Reference trajectories are tested and verified in simulation as well as on a test stand.

关键词： Vibrations Productivity Mechatronics Finite impulse response filters Kinematics filtering algorithms Trajectory

来源：评论

学校读者我要写书评

暂无评论

Dynamic Variable Step Size LMS Adaptation algorithms—Application to Adaptive Feedforward Noise Attenuation*

Dynamic Variable Step Size LMS Adaptation Algorithms—Applic...

引用

European Control Conference (ECC)

作者： Tudor-Bogdan Airimitoaie Bernard Vau Dariusz Bismor Gabriel Buche Ioan Doré Landau Univ. Bordeaux CNRS Bordeaux INP IMS Talence France Exail Bonneuil sur Marne France Silesian University of Technology Gliwice Poland GIPSA-lab Univ. Grenoble Alpes CNRS Grenoble INP Grenoble France

ISBN: (数字)9783907144107

ISBN: (纸本)9798331540920

The paper explores in detail the use of dynamic adaptation gain/step size (DAG) for improving the adaptation transient performance of variable step-size LMS (VS-LMS) adaptation algorithms. A generic form for the implementation of the DAG within the VS-LMS algorithms is provided. Criteria for the selection of the coefficients of the DAG filter which is required to be a strictly positive real transfer operator are given. The potential of the VS-LMS adaptation algorithms using a DAG is then illustrated by experimental results obtained on a relevant adaptive active noise attenuation system.

关键词： Adaptive systems Heuristic algorithms Noise Transfer functions Europe filtering algorithms Attenuation

来源：评论

学校读者我要写书评

暂无评论

Drone-Assisted QR Code Recognition Method for Warehouse Management

Drone-Assisted QR Code Recognition Method for Warehouse Mana...

引用

IEEE International Conference on Networking, Sensing and Control

作者： Beiwen Chen Jiaqi Huang Faculty of Innovation Engineering Macau University of Science and Technology Macau China

ISBN: (数字)9798350365221

ISBN: (纸本)9798350365238

In response to the increasing complexity of inventory management in large-scale warehouses, we present a drone-assisted quick response (QR) code recognition technology for warehouse management. It combines panoramic stitching and image preprocessing for QR code recognition. The experiments are conducted and performed in two phases. First, we compare three basic QR code recognition methods based on Pyzbar, OpenCV, and dynamsoft barcode reader (DBR), with experimental results indicating that DBR performs the best. Second, building on the DBR method, we compare the recognition effects of different image preprocessing techniques, namely the method proposed in this paper, geometric correction, sharpening (SH) and adaptive histogram equalization (AHE), and the median filter algorithm. The results show that the proposed method excels in average processing time per batch (τ), stitching success rate (ŋ), and the number of recognized QR codes (ξ). The average processing time per batch is 3.13 seconds, the stitching success rate (ŋ) is 63.43%, and totally 746 QR codes are recognized. It approaches the performance of geometric correction methods with its advantage in processing speed. It provides significant practical value and innovative solutions for enhancing efficiency in modern warehouse management.

关键词： Image recognition Machine learning algorithms Accuracy QR codes filtering algorithms Robustness Sensors Image preprocessing Distributed Bragg reflectors Proposals

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：