检索结果-内蒙古大学图书馆

Half space object classification via incident angle based fusion of radar and infrared sensors

Journal of systems engineering and Electronics 2022年第5期33卷 1025-1031页

作者： HE Zhenyu ZHUGE Xiaodong WANG Junxiang YU Shihao XIE Yongjun ZHAO Yuxiong School of Electronic Information and Engineering Beihang UniversityBeijing 100191China Beijing Institute of Electronic System Engineering Beijing 100854China Beijing Institute of Astronautical Systems Engineering Beijing 100076China Shenzhen Institute of Beihang University Shenzhen 518038China School of Software Beihang UniversityBeijing 100191China

In this paper,we introduce an incident angle based fusion method for radar and infrared sensors to improve the recognition rate of complex targets under half space scenarios,e.g.,vehicles on the ground in this *** radar sensors,convolutional operation is introduced into the autoencoder,a“winner-take-all(WTA)”convolutional autoencoder(CAE)is used to improve the recognition rate of the radar high resolution range profile(HRRP).Moreover,different from the free space,the HRRP in half space is more *** order to get closer to the real situation,the half space HRRP is simulated as the *** recognition rate has a growth more than 7%com-pared with the traditional CAE or denoised sparse autoencoder(DSAE).For infrared sensor,a convolutional neural network(CNN)is used for infrared image ***,we com-bine the two results with the Dempster-Shafer(D-S)evidence theory,and the discounting operation is introduced in the fusion to improve the recognition *** recognition rate after fusion has a growth more than 7%compared with a single *** the discounting operation,the accuracy rate has been improved by 1.5%,which validates the effectiveness of the proposed method.

关键词： convolutional autoencoder(CAE) half space high-resolution range profile(HRRP) incident angle based fusion tar-get recognition

来源：评论

学校读者我要写书评

暂无评论

Reinforcement Learning for Imbalanced Vehicle Booming Noise Classification 2

Reinforcement Learning for Imbalanced Vehicle Booming Noise ...

引用

2nd International Conference on Machine Learning, Control, and Robotics, MLCR 2023

作者： Yang, Jenny Kunte, Deepti S. Cornelis, Bram Clifton, David A. University of Oxford Institute of Biomedical Engineering Dept. of Engineering Science Oxford United Kingdom Siemens Digital Industries Software Nv Leuven Belgium nic System Dynamics Dept. of Mechanical Engineering Leuven Belgium

ISBN: (纸本)9798350381245

Booming noise, along with other factors contributing to interior vehicle noise, significantly influences the overall perception of an automobile's quality. In light of the automotive industry's strong focus on customer satisfaction, developing robust sound quality prediction models and comprehensive testing procedures is of paramount importance. These endeavors assist manufacturers in delivering a more enjoyable and comfortable driving experience, ultimately strengthening their vehicles' reputation in the market. However, the datasets used for training sound quality prediction models often exhibit class imbalances, posing a substantial challenge for machine learning. This imbalance can result in diminished predictive performance, especially for the minority class. Booming noise detection is particularly sensitive to this issue, given the greater prevalence of normal conditions compared to faulty ones. In our study, we address imbalanced learning across three vehicle types and investigate various imbalance ratios. Our findings highlight the effectiveness of a novel reinforcement learning framework in enhancing model generalizability and robustness to noise in the context of imbalanced data, outperforming traditional methods in these aspects. © 2023 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

MSE-CapsPPISP: Spatial Hierarchical Protein-Protein Interaction Sites Prediction Using Squeeze-and-Excitation Capsule Networks

MSE-CapsPPISP: Spatial Hierarchical Protein-Protein Interact...

引用

2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022

作者： Lv, Weipeng Jiang, Changkun Li, Jianqiang Shenzhen University College of Computer Science and Software Engineering National Engineering Laboratory for Big Data System Computing Technology Shenzhen China

ISBN: (纸本)9781665468190

The discovery of protein-protein interaction sites (PPIs) is vital for exploring the principle of PPI and understanding the nature of life activities. Developing computational approaches to predict PPIs can effectively compensate for the shortcomings of biological experiments, which are mostly time-consuming and vulnerable to noise. In recent years, deep learning has been used to develop PPIs prediction models. Most of them consider the contextual information of the target amino acid residues and use a local protein sequence to represent the targets. However, the traditional deep-learning techniques, e.g., deep neural networks (DNNs) and convolutional neural networks (CNNs), disregard the important spatial hierarchies contained in the features of protein sequences, leading to their failure to effectively distinguish the interaction sites from different residue regions. In this work, we design MSE-CapsPPISP, a new deep-learning model to address the PPIs prediction with spatial hierarchies. The key idea of MSE-CapsPPISP is to take into account the hierarchical relationships between the features of protein sequences. We characterize the hierarchical relationships by designing a tailored Capsule Network (CapsNet) model, which is a novel type of neural network with vector neurons. Moreover, to make the network representation more robust, MSE-CapsPPISP uses multi-scale CNNs to extract multi-scale features of protein sequences and Squeeze-and-Excitation blocks to recalibrate the features. The validation results show that our MSE-CapsPPISP outperforms the baseline CNNs-based architecture DeepPPISP and other competing schemes in the PPIs prediction task. © 2022 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Preliminary denoising by 3D U-Net in image domain for low dose CT images 2022

Preliminary denoising by 3D U-Net in image domain for low do...

引用

2nd International Conference on Bioinformatics and Intelligent Computing, BIC 2022

作者： Song, Xiaofu Han, Yu Xi, Xiaoqi Li, Lei Zhu, Linlin Yang, Shuangzhan Liu, Mengnan Tan, Siyu Yan, Bin Zhengzhou University School of Software Zhengzhou45000 China Information Engineering University College of Information System Engineering Zhengzhou45000 China

ISBN: (纸本)9781450395755

Low dose CT (LDCT) by reducing the X-ray tube current is of huge significance during clinical scanning. However, low-dose CT images often have strong noise and artifacts, which affects the image quality and diagnostic performance. LDCT noise reduction methods based on deep learning have recently achieved good results in improving image quality. Since the reconstructed CT image itself is 3D, in this paper a LDCT denoising method based on 3D U-Net is proposed to combine the 3D spatial information by 3D convolution directly, instead of processing 2D slices from 3D volume data. Therefore, the image change continuity between the adjacent slices is guaranteed. In addition, multiple down-sampling operations in the network, which can reduce the number of parameters of the 3D network, help the network to train. The experimental results show that the proposed method can effectively preserve the structural and texture information of normal NDCT images and significantly suppress the image noise and artifacts, achieving better performance in both quantification and visualization. Compared with LDCT images without denoising, the peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) of the processed images were improved by 12.18 dB and 0.35 dB, respectively. © 2022 ACM.

关键词： Computerized tomography

来源：评论

学校读者我要写书评

暂无评论

Evolutionary Multiobjective Feature Selection Assisted by Unselected Features 13

Evolutionary Multiobjective Feature Selection Assisted by Un...

引用

13th IEEE Congress on Evolutionary Computation, CEC 2024

作者： Duan, Xuan Liu, Songbai Ji, Junkai Li, Lingjie Lin, Qiuzhen Tan, Kay Chen College of Computer Science and Software Engineering Shenzhen University Shenzhen China Shenzhen University National Engineering Laboratory for Big Data System Computing Technology Shenzhen China Shenzhen China The Hong Kong Polytechnic University Department of Computing Hong Kong

ISBN: (纸本)9798350308365

To enhance the generalization of multi-objective feature selection (MOFS) in classification, this paper proposes an evolutionary multitasking algorithm, diverging from previous approaches that exclusively target selected features. The algorithm integrates information from both selected and unselected features, introducing a novel objective to minimize the accuracy of unselected features. This objective, combined with the goal of minimizing classification errors for selected features, forms an auxiliary MOFS task. The paper presents a dual-population evolutionary multitasking framework that synergizes the main MOFS task with the auxiliary task. A knowledge transfer mechanism, based on accuracy preferences, seamlessly shares insights from the auxiliary to the main task, aiming to identify improved Pareto feature subsets. Empirical results demonstrate the superior performance of several state-of-the-art multi-objective algorithms within this framework, highlighting significant improvements across diverse datasets. © 2024 IEEE.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

Optimized Evaluation of Mobile Base Station by Modern Topological Invariants

引用

Computers, Materials & Continua 2023年第1期74卷 363-378页

作者： Khalid Hamid Muhammad Waseem Iqbal Muhammad Usman Ashraf Ahmed Mohammed Alghamdi Adel A.Bahaddad Khalid Ali Almarhabi Department of Computer Science Superior UniversityLahore54000Pakistan Department of Software Engineering Superior UniversityLahore54000Pakistan Department of Computer Science GC Women UniversitySialkot53310Pakistan Department of Software Engineering College of Computer Science and EngineeringUniversity of Jeddah21493Saudi Arabia Department of Information System Faculty of Computing and Information TechnologyKing Abdulaziz UniversityJeddah21589Saudi Arabia Department of Computer Science College of Computing in Al-QunfudahUmm Al-Qura UniversityMakkah24381Saudi Arabia

Due to a tremendous increase in mobile traffic,mobile operators have started to restructure their networks to offload their *** directions will lead to fundamental changes in the design of future Fifthgeneration(5G)cellular *** the formal reason,the study solves the physical network of the mobile base station for the prediction of the best characteristics to develop an enhanced network with the help of graph *** number that can be uniquely calculated by a graph is known as a graph *** the last two decades,innumerable numerical graph invariants have been portrayed and used for correlation *** any case,no efficient assessment has been embraced to choose,how much these invariants are connected with a network *** paper will talk about two unique variations of the hexagonal graph with great capability of forecasting in the field of optimized mobile base station topology in setting with physical *** K-banhatti sombor invariants(KBSO)and Contrharmonic-quadratic invariants(CQIs)are newly introduced and have various expectation characteristics for various variations of hexagonal graphs or *** the hexagonal networks are used in mobile base stations in layered,forms called *** review settled the topology of a hexagon of two distinct sorts with two invariants KBSO and CQIs and their reduced *** deduced outcomes can be utilized for the modeling of mobile cellular networks,multiprocessors interconnections,microchips,chemical compound synthesis and memory interconnection *** results find sharp upper bounds and lower bounds of the honeycomb network to utilize the Mobile base station network(MBSN)for the high load of traffic and minimal traffic also.

关键词： Hexagonal networks k-banhatti sombor invariants maple network graph artificial intelligence contraharmonic-quadratic invariants honeycomb

来源：评论

学校读者我要写书评

暂无评论

An improved scattering correction method for model analysis of cone-beam CT 2022

An improved scattering correction method for model analysis ...

引用

2nd International Conference on Bioinformatics and Intelligent Computing, BIC 2022

作者： Kang, Guanyu Han, Yu Xi, Xiaoqi Li, Lei Yang, Shuangzhan Zhu, Linlin Liu, Mengnan Tan, Siyu Yan, Bin Zhengzhou University School of Software Zhengzhou45000 China Information Engineering University College of Information System Engineering Zhengzhou45000 China

ISBN: (纸本)9781450395755

Due to the increased area of X-ray interaction with the objects, the cone beam computed tomography (CBCT) systems will bring severe scattering artifacts in reconstructed images, and the correction of which is quite essential to a good image quality. The scatter process modeling methods without Monte Carlo simulation proposed in recent years is relatively efficient to obtain the scattering distribution. In this paper, combined with Compton scattering cross section (K-N) formula, an improved scatter correction method is proposed for single-material objects. After setting the scattering points in the object from the reconstruction image, it is derived that the scattering ratio of any two scattering points at scattering angle 0 ° is approximately equal to the projection ratio of the object on the corresponding path of two scattering points. Therefore, the first-order Compton scattering values of all scattering points at scattering angle 0 ° can be calculated. The scattering probability of the same scattering point at different scattering angles are then normalized to obtain the scattering ratio. The scattering distribution of each scattering point was calculated by combining the scattering value at 0 ° and the included angle and distance between the scattering point and the detector. Since the attenuation coefficient selection and the attenuation path length calculation before the scattering points are omitted, the scattering estimation error is reduced, and the scattering estimation efficiency is improved. The experimental correction results for the CT images of the aircraft aluminum castings show that the proposed method can effectively restore the structural information of scattering effects, enhancing the contrast to noise ratio (CNR) and reducing the cup shadow from 9.07% to 5.49%. © 2022 ACM.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

EagerLog: Active Learning Enhanced Retrieval Augmented Generation for Log-based Anomaly Detection

EagerLog: Active Learning Enhanced Retrieval Augmented Gener...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Duan, Chiming Jia, Tong Yang, Yong Liu, Guiyang Liu, Jinbu Zhang, Huxing Zhou, Qi Li, Ying Huang, Gang School of Software and Microelectronics Peking University Beijing China Institute for Artificial Intelligence Peking University Beijing China National Engineering Research Center for Software Engineering Peking University Beijing China National Key Laboratory of Data Space Technology and System Beijing China Alibaba Cloud Computing Company Hangzhou China

ISBN: (纸本)9798350368741

Logs record essential information about system operations and serve as a critical source for anomaly detection, which has generated growing research interest. Utilizing large language models (LLMs) within a retrieval-augmented generation (RAG) framework for log-based anomaly detection is an effective approach due to its strong generalization capabilities and efficient few-shot performance. However, the effectiveness of this method hinges on the quality of the knowledge source, which can be impacted by noise and changes within the software systems. Facing these problems, in this paper, we propose a novel log-based anomaly detection method named EagerLog, employing active learning to choose the logs for humans to label, thereby adding them to the knowledge source, thus enhancing the knowledge source and maintaining its quality. Our experiments on three open datasets (BGL, Thunderbird, Zookeeper) and one industrial dataset demonstrate that EagerLog can achieve 93.65% F1 score with approximately 10 labeled log sequences, surpassing existing methods by 15.32%. © 2025 IEEE.

关键词： Active Learning Anomaly Detection Log Analysis Retrieval Augmented Generation

来源：评论

学校读者我要写书评

暂无评论

A reliable method for data aggregation on the industrial internet of things using a hybrid optimization algorithm and density correlation degree

引用

Cluster Computing 2024年第6期27卷 7521-7539页

作者： Heidari, Arash Shishehlou, Houshang Darbandi, Mehdi Navimipour, Nima Jafari Yalcin, Senay Department of Software Engineering Haliç University Istanbul34060 Turkey Department of Computer Engineering Tabriz Branch Islamic Azad University Tabriz Iran Pôle Universitaire Léonard de Vinci Paris France Department of Computer Engineering Kadir Has University Istanbul Turkiye and Future Technology Research Center National Yunlin University of Science and Technology Yunlin Douliou64002 Taiwan Department of Energy System Engineering School of Engineering and Natural Sciences Bahçeşehir University Istanbul Turkey

The Internet of Things (IoT) is a new information technology sector in which each device may receive and distribute data across a network. Industrial IoT (IIoT) and related areas, such as Industrial Wireless Networks (IWNs), big data, and cloud computing, have made significant strides recently. Using IIoT requires a reliable and effective data collection system, such as a spanning tree. Many previous spanning tree algorithms ignore failure and mobility. In such cases, the spanning tree is broken, making data delivery to the base station difficult. This study proposes an algorithm to construct an optimal spanning tree by combining an artificial bee colony, genetic operators, and density correlation degree to make suitable trees. The trees’ fitness is measured using hop count distances of the devices from the base station, residual energy of the devices, and their mobility probabilities in this technique. The simulation outcomes highlight the enhanced data collection reliability achieved by the suggested algorithm when compared to established methods like the Reliable Spanning Tree (RST) construction algorithm in IIoT and the Hop Count Distance (HCD) based construction algorithm. This proposed algorithm shows improved reliability across diverse node numbers, considering key parameters including reliability, energy consumption, displacement probability, and distance. © The Author(s) 2024.

关键词： Reliability

来源：评论

学校读者我要写书评

暂无评论

Deep Learning to Classify Bacterial Species in the same Genus

Deep Learning to Classify Bacterial Species in the same Genu...

引用

2024 International Conference on Electronics, Information, and Communication, ICEIC 2024

作者： Sheela, Sherin Piang, May Phu Sakorntanant, Sakda Poomrittigul, Suvit Treebupachatsakul, Treesukon School of Engineering King Mongkut's Institute of Technology Ladkrabang Department of Biomedical Engineering Thailand Pathumwan Institute of Technology Faculty of Science and Technology Department of Software Engineering and Information System Bangkok Thailand School of Information Technology King Mongkut's Institute of Technology Ladkrabang Thailand

ISBN: (纸本)9798350371888

Bacterial strains in the same genus share highly similar morphology, gram-staining characteristics, colony sizes, and spatial arrangements. Therefore, identifying them by deep learning can be quite challenging. This study aimed to assess the classification of 7 species of bacteria from 2 genera of Bacillus and Vibrio by using 8 Convolutional Neural Network (CNN) models. We implemented Python programming along with Keras API within the Jupyter Notebook. The models were constructed and evaluated under unbalanced and balanced datasets by augmentation (rotation, flip, etc.). Transfer learning with fine-tuning, and pre-processing of mixup and label smoothing were also applied to reduce overfitting and enhance generalization. Based on the experimental results on private dataset, the results of InceptionResNetV2 emerged as the top-performing model with a notable accuracy of 82.8%, 88.6% precision, 78.4% recall, and 78.0% F1-score when label smoothing was applied at 0.5 on balanced dataset. © 2024 IEEE.

关键词： Neural network models

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：