检索结果-内蒙古大学图书馆

Monaural speech enhancement using U-net fused with multi-head self-attention

Chinese Journal of Acoustics 2023年第1期42卷 98-118页

作者： FAN Junyi YANG Jibin ZHANG Xiongwei ZHENG Changyan Graduate School Army Engineering UniversityNanjing 210007 College of Command and Control Engineering Army Engineering UniversityNanjing 210007 Department of Test Control High-tech InstituteQingzhou 262500

Under low signal-to-noise ratio(SNR)and burst noise conditions,the speech enhancement effect of existing deep learning network models is not *** contrast,humans can exploit the long-term correlation of speech to form an integrated perception of different speech ***,describing the long-term dependencies of speech can help improve the enhancement performance under low SNR and burst noise *** by this feature,a time domain end-to-end monaural speech enhancement model TU-net that fuses the multi-head self-attention mechanism and U-net deep network is *** TU-net model adopts the codec layer structure of U-net to achieve multi-scale feature *** introduces the dual-path Transformer module using the multi-head self-attention mechanism to calculate the speech mask and better model long-term *** TU-net model is trained with a weighted sum loss function in the time,time-frequency,and perceptual *** experiments are carried out and the results show that with maintaining relatively fewer network model parameters,TU-net outperforms other similar monaural enhancement network models in several evaluation metrics such as perceptual evaluation of speech quality(PESQ),short-time objective intelligibility(STOI)and SNR gain under low SNR and burst noise conditions.

关键词： network speech noise

来源：评论

学校读者我要写书评

暂无评论

Aerial refueling scheduling of multi-receiver and multi-tanker under spatial-temporal constraints for forest firefighting

引用

Chinese Journal of Aeronautics 2024年第5期37卷 71-91,I0001页

作者： Jing HUANG Jiaqi XING Jinrui REN Quan QUAN Youmin ZHANG Department of Information and Control Engineering Xi’an University of TechnologyXi’an 710048China Jiujiang Measuring Test Technology Research Institute Jiujiang 332000China School of Automation Science and Electrical Engineering Beihang UniversityBeijing 100191China School of Automation Xi’an University of Posts&TelecommunicationsShaanxiXi’an 710121China Department of Mechanical Industrial and Aerospace EngineeringConcordia UniversityMontreal H3G 1M8Canada

Forest fires pose a significant threat to human life and property,so the utilization of unmanned aircraft systems provides new ways for forest *** the constrained load capacities of these aircraft,aerial refueling becomes crucial to extend their operational time and *** order to address the complexities of firefighting missions involving multi-receiver and multi-tanker deployed from various airports,first,a fuel consumption calculation model for aerial refueling scheduling is established based on the receiver ***,two distinct methods,including an integrated one and a decomposed one,are designed to address the challenges of establishing refueling airspace and allocating tasks for *** methods aim to optimize total fuel consumption of the receivers and tankers within the aerial refueling scheduling *** optimization problem is established as nonlinear optimization models along with *** integrated method seamlessly combines refueling rendezvous point scheduling and tanker task allocation into unified *** has a complete solution space and excels in optimizing total fuel *** decomposed method,through the separation of rendezvous point scheduling and task allocation,achieves a reduced computational ***,this comes at the cost of sacrificing optimality by excluding specific feasible ***,numerical simulations are carried out to verify the feasibility and effectiveness of the proposed *** simulations yield insights crucial for the practical engineering application of both the integrated and decomposed methods in real-world *** comprehensive approach aims to enhance the efficiency of forest firefighting operations,mitigating the risks posed by forest fires to human life and property.

关键词： Aerial refueling Scheduling Planning Spatial-temporal constraint Firefighting

来源：评论

学校读者我要写书评

暂无评论

Astronomical Image Processing by the Lemur Software 24

Astronomical Image Processing by the Lemur Software

引用

6th Asia Conference on Machine Learning and Computing, ACMLC 2024

作者： Khlamov, Sergii Savanevych, Vadym Briukhovetskyi, Oleksandr Vlasenko, Vladimir Dikov, Yevhen Dmytrenko, Artem Department of Media Systems and Technologies Kharkiv National University of Radio Electronics Kharkiv Ukraine Department of systems engineering Kharkiv National University of Radio Electronics Kharkiv Ukraine Western Center of Radio Engineering Surveillance National Space Facilities Control and Test Center Kyiv Ukraine Space Research and Communication Center National Space Facilities Control and Test Center Kyiv Ukraine Department of Software Engineering GlobalLogic Incorporation Kharkiv Ukraine Institute of Astronomy V. N. Karazin Kharkiv National University Kharkiv Ukraine

ISBN: (纸本)9798400710018

The astronomical images are made by cameras with the charge-coupled device (CCD). They can be received from the different sources, like servers, clusters, predefined series, archives or "live" (online) data streams. The astronomical images processing is focused on but not limited to the following tasks: data mining, knowledge discovery, big astronomical data processing, filtering, background alignment, brightness equalization, segmentation, classification, image recognition, object's image detection, object's astrometry and photometry, moving object detection, parameters determination of the object's image and apparent motion, reference objects selection and others. The modern Lemur software of the Collection Light Technology (CoLiTec) project (https://***) was developed using the described above technologies and approaches. The Lemur software is designed to perform a sequence of the following main steps: pre-processing (astronomical information collection -> worst data rejection -> useful data extraction -> data mining -> classification -> background alignment -> brightness equalization), image processing (segmentation -> typical form analysis -> recognition patterns applying -> detection of the object's image -> astrometry -> photometry -> objects identification -> tracks detection), knowledge discovery (Solar System objects or artificial satellites to be discovered, tracks parameters for the investigation, light curves of the variable stars, scientific reports in the international formats). The paper describes the modern features for the astronomical image processing implemented in the Lemur software. It has assisted in making over 1700 discoveries of asteroids, including 5 NEOs, 21 Trojan asteroids of Jupiter, 1 Centaur. In total it has been used in about 800 000 observations, during which five comets were discovered. All such observations and discoveries have official observational circulares from the Minor Planet Center (MPC) of the International Ast

关键词： Asteroids

来源：评论

学校读者我要写书评

暂无评论

Class Incremental Learning Method for Hyperspectral Images Based on Real Data Playback Mechanism and Classification Network Optimization 15th

Class Incremental Learning Method for Hyperspectral Images B...

引用

15th International Conference on Genetic and Evolutionary Computing, ICGEC 2023

作者： Liu, Xiaodong Wang, Siqi Liu, Huanyu Li, Junbao Department of Automatic Test and Control School of Electronics and Information Engineering Harbin Institute of Technology Harbin150080 China School of Computer Science and Technology Information Countermeasure Technique Institute Harbin Institute of Technology Harbin150080 China

ISBN: (纸本)9789819700677

In recent years, the hyperspectral image (HSI) classification has attracted great attention in the field of earth observation. With the expansion of application scenarios and the continuous improvement of application requirements, new classes of HSI continue to emerge. In the face of open and dynamic application scenarios, the model is required to be able to continuously learn new categories based on maintaining existing category knowledge. Class incremental learning algorithms have received extensive attention as key solutions to the problem above. In this paper, we use a regularization-based incremental learning algorithm, elastic weight consolidation (EWC) for class incremental learning for HSI classification. After data preprocessing to reduce data dimensions, we select new classes for research on class incremental learning. We combine 3D-2D convolutional structures and residual blocks as the classification network and use real data playback mechanism to improve class recognition accuracy and realize class incremental recognition of HSI data. We demonstrate the feasibility of the algorithm with extensive experiments based on three widely used hyperspectral datasets. © 2024, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Learning algorithms

来源：评论

学校读者我要写书评

暂无评论

High-Frequency Filtration for the Brightness Alignment of the Astronomical Frame Background 14

High-Frequency Filtration for the Brightness Alignment of th...

引用

14th International Conference on Advanced Computer Information Technologies, ACIT 2024

作者： Vlasenko, Vladimir Khlamov, Sergii Savanevych, Vadym Trunova, Tetiana Tabakova, Iryna Mendeleva, Mariia National Space Facilities Control and Test Center Space Research and Communication Center Kyiv Ukraine Kharkiv National University of Radio Electronics Department of Media Systems and Technologies Kharkiv Ukraine Kharkiv National University of Radio Electronics Department of Systems Engineering Kharkiv Ukraine

ISBN: (纸本)9798350350036

This paper explores the significance of image processing in various fields of scientific research, including astronomy, where it plays a crucial role in enhancing the quality of acquired data and assists researchers in extracting valuable information from them. One important aspect of image processing is brightness equalization, which improves the contrast and sharpness of images, making them more suitable for analysis and interpretation. For instance, in astronomical observations, this can be critical for distinguishing faint objects against the background of the night sky or galaxies. The study also examines various high-frequency filtering methods, such as the Gaussian filter, Butterworth filter, and ideal filter, and their application in the context of astronomical observations. These methods play a vital role in processing astronomical images by helping to eliminate noise and artifacts that may distort the data. In astronomical images, noise can be caused by atmospheric effects, electronic interference, or even cosmic rays. The application of appropriate filtering methods reduces the influence of these factors and enhances the reliability of the obtained results. Thus, image processing is an integral part of modern astronomy, playing a significant role in acquiring and analyzing data about cosmic objects and phenomena. © 2024 IEEE.

关键词： Butterworth filters

来源：评论

学校读者我要写书评

暂无评论

Study on Target Detection and Tracking Method of UAV Based on LiDAR 12

Study on Target Detection and Tracking Method of UAV Based o...

引用

12th IEEE Global Reliability and Prognostics and Health Management, PHM-Nanjing 2021

作者： Wang, Hong Peng, Yu Liu, Liansheng Liang, Jun Harbin Institute Of Technology Department Of Test And Control Engineering Harbin China

ISBN: (纸本)9781665401302

Unmanned aerial vehicle (UAV) may bring potential danger such as hurting people falling accidents, Therefore the surveillance of trajectory of UAV is subjected to extensive attention for avoiding irreversible damage. This paper presents a target detection and tracking method based on LiDAR, in which target detection uses Euclidean distance clustering and particle filter are utilized to achieve target detection and tracking, respectively. Once UAV is detected by the clustering method, the particle filter algorithm is used to track the target. In the experiment, the Livox Mid 40 LiDAR developed by the Livox company is used as the front-end sensor to detect the UAV. Experimental results show that the proposed method could realize the detection and tracking of UAV. © 2021 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Optical Stealth Algorithm Under Photoelectric Countermeasures and Its Effectiveness Evaluation 18th

Optical Stealth Algorithm Under Photoelectric Countermeasure...

引用

18th International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2022

作者： Yang, Zhonglin Jiang, Yutong Li, Junbao Li, Qiong Department of Automatic Test and Control School of Electronics and Information Engineering Harbin Institute of Technology Harbin150080 China China North Vehicle Research Institute 4 HuaiShuling Fengtai District Beijing China

ISBN: (纸本)9789819901043

With the diversification of reconnaissance equipment and methods in the military field, optical stealth technology has become a hot spot in military research under the operational needs of optoelectronic confrontation. An improved self-adaptive visible light and infrared waveband stealth algorithm is proposed here, and its effectiveness under photoelectric confrontation is evaluated. The current digital camouflage design method is mainly through background color mixing, image segmentation, spot combination, etc. It has the defects of smooth and clear edge contour, fixed color/texture scene, and cannot adapt to complex texture background or multiple types of scenes. An improved digital camouflage design method based on fuzzy C-means clustering (FCM) image processing technology is proposed to obtain the color characteristics of the background image. Then, the color block types of the camouflage spots are acquired to reduce the loss of texture information of the background image caused by clustering or segmentation to avoid the phenomenon of clear boundaries caused by single region segmentation. The results show that the use of adaptive optical stealth algorithms in the visible and infrared bands can reduce the probability of being discovered in optoelectronic countermeasures, thereby gaining a head start. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Nonlinear Modeling of Fixed-Wing Mode for a Ducted-Fan Tiltrotor UAV by DATCOM

Nonlinear Modeling of Fixed-Wing Mode for a Ducted-Fan Ti...

引用

International Conference on Guidance, Navigation and control, ICGNC 2022

作者： Huang, Jing Wang, Ban Deng, Yangping Zhang, Youmin Department of Information and Control Engineering Xi’an University of Technology Xi’an China Jiujiang Measuring Test Technology Research Institute Jiujiang China School of Aeronautics Northwestern Polytechnical University Xi’an China Department of Mechanical Industrial and Aerospace Engineering Concordia University Montreal Canada

ISBN: (纸本)9789811966125

The VTOL UAVs draw more and more attention from research institute recently. This paper proposes a nonlinear modeling process of fixed-wing mode for a novel tiltrotor unmanned aerial vehicle with double ducted fans by Digital DATCOM and Simulink. With the help of the above two simulation software, the flight vehicle design process is simplified and accelerated. Firstly, a new ducted tiltrotor UAV design scheme with the capabilities of VTOL and high velocity cruise is presented. Then, the coefficients of proposed ducted tiltrotor UAV are calculated by Digital DATCOM, and the model of vehicle is assembled in Simulink. Finally, simulation result shows that the vehicle is trimming at the condition point of velocity 20.4 m/s, and it realizes straight level flight in this point. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Fixed wings

来源：评论

学校读者我要写书评

暂无评论

Design and Realization of Unmanned Vehicle Autonomous Navigation System Based on Binocular Camera 3

Design and Realization of Unmanned Vehicle Autonomous Naviga...

引用

3rd International Conference on Computer Science, Electronic Information engineering and Intelligent control Technology, CEI 2023

作者： Guo, Tianxing Lin, Lianlei Liu, Huanyu Li, Junbao School of Electronics and Information Engineering Harbin Institute of Technology Department of Automatic Test and Control Harbin China Information Countermeasure Technique Institute School of Computer Science and Technology Harbin Institute of Technology Harbin China

ISBN: (纸本)9798350372052

Aiming at the navigation problem of unmanned vehicles in extreme environments such as communication interference and limited GPS signals, this study proposes an autonomous navigation method based on binocular cameras. The method enables the unmanned vehicle to complete the task of localization and map building using only binocular images, without the need for other sensors or signal sources. Meanwhile, this study also proposes a local path planning obstacle avoidance method based on depth map, which, when combined with global localization information, can take into account both local obstacle avoidance and global path navigation, and can cope with random environmental changes. The system uses a single sensor for autonomous obstacle avoidance and navigation, which reduces the computational requirements and ensures the low cost of the navigation system. In order to verify the reliability and effectiveness of the system under signal-constrained conditions, the system is evaluated in a simulation environment and a real field scenario, respectively. The experimental results show that the system is able to achieve reliable localization and path planning under signal-constrained conditions. © 2023 IEEE.

关键词： Cameras

来源：评论

学校读者我要写书评

暂无评论

Short‐term and long‐term memory self‐attention network for segmentation of tumours in 3D medical images

引用

CAAI Transactions on Intelligence Technology 2023年第4期8卷 1524-1537页

作者： Mingwei Wen Quan Zhou Bo Tao Pavel Shcherbakov Yang Xu Xuming Zhang Department of Biomedical Engineering College of Life Science and TechnologyHuazhong University of Science and TechnologyWuhanChina State Key Laboratory of Digital Manufacturing Equipment and Technology Huazhong University of Science and TechnologyWuhanChina Institute for Control Science Russian Academy of SciencesMoscowRussia Hubei Medical Devices Quality Supervision and Test Institute WuhanChina

Tumour segmentation in medical images(especially 3D tumour segmentation)is highly challenging due to the possible similarity between tumours and adjacent tissues,occurrence of multiple tumours and variable tumour shapes and *** popular deep learning‐based segmentation algorithms generally rely on the convolutional neural network(CNN)and *** former cannot extract the global image features effectively while the latter lacks the inductive bias and involves the complicated computation for 3D volume *** existing hybrid CNN‐Transformer network can only provide the limited performance improvement or even poorer segmentation performance than the pure *** address these issues,a short‐term and long‐term memory self‐attention network is ***,a distinctive self‐attention block uses the Transformer to explore the correlation among the region features at different levels extracted by the ***,the memory structure filters and combines the above information to exclude the similar regions and detect the multiple ***,the multi‐layer reconstruction blocks will predict the tumour *** results demonstrate that our method outperforms other methods in terms of subjective visual and quantitative *** with the most competitive method,the proposed method provides Dice(82.4%vs.76.6%)and Hausdorff distance 95%(HD95)(10.66 vs.11.54 mm)on the KiTS19 as well as Dice(80.2%vs.78.4%)and HD95(9.632 vs.12.17 mm)on the LiTS.

关键词： 3D medical images convolutional neural network self‐attention network transformer tumor segmentation

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：