检索结果-内蒙古大学图书馆

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

作者： Hongming Fu Guanyao Wu Zhu Liu Tiantian Yan Jinyuan Liu School of Software Technology Dalian University of Technology Dalian China National and Local Joint Engineering Laboratory of Computer Aided Design School of Software Engineering Dalian University Dalian China

A series of infrared and visible image fusion (IVIF) methods have emerged to improve the performance of segmentation task. However, existing perception-focused IVIF methods take visual effects and semantic information as a unified goal for training, ignoring the task conflicts. Moreover, these methods often involve manually designed modules, which are laborious and suboptimal. To solve the problems, we propose a collaborative feature learning framework based on neural architecture search (NAS). Specifically, we extract shared features of fusion and segmentation tasks into a unified space and separately process task objectives through a dual decoder. In light of the essential role that semantic information plays in the segmentation task, we construct a hybrid search space with transformers incorporated to enhance context dependence handling. Our method undergoes extensive experiments, showcasing exceptional visual effects and significant enhancements in segmentation tasks compared to other state-of-the-art methods.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Comparative analysis of CM EMI for boost converter using Si/SiC hybrid switch 1

Comparative analysis of CM EMI for boost converter using Si/...

引用

1st IEEE China International Youth Conference on electrical engineering, CIYCEE 2020

作者： Hu, Bo Peng, Zishun Xiao, Pei Dai, Yuxing Zhang, Chao Wang, Jun Hunan University College of Electrical and Information Engineering China Wenzhou University National-Local Joint Engineering Laboratory of Digitalize Electrical Design Technology China

ISBN: (纸本)9781728196596

The combination of the Si IGBT and SiC MOSFET inside the Si/SiC hybrid switch will bring serious conducted electromagnetic interference (EMI) noise in the application of the power converters. Compared with the differential mode (DM) EMI noise, the common mode (CM) EMI noise has a greater impact on the total conducted EMI noise, and it is relatively difficult to suppress the common mode EMI noise. However, there is still lack of research on the CM EMI emission characteristics of Si/SiC hybrid switch based converters. In this paper, the CM EMI noise caused by the Si/SiC hybrid switch, Si IGBT, and SiC MOSFET are compared at first. Then, the influence of different current ratios of the Si/SiC hybrid switch on the CM EMI is analyzed. © 2020 IEEE.

关键词： Silicon carbide

来源：评论

学校读者我要写书评

暂无评论

Aφ6-m Tunnel Boring Machine Steel Arch Splicing Manipulator

引用

Chinese Journal of Mechanical engineering 2022年第2期35卷 121-134页

作者： Yuanfu He Yimin Xia Zhen Xu Jie Yao Bo Ning Xuemeng Xiao College of Mechanical and Electrical Engineering Central South UniversityChangsha 410083China State Key Laboratory of High Performance Complex Manufacturing Central South UniversityChangsha 410083China China Railway Construction Heavy Industry Co.Ltd. Changsha 410100China China Railway Siyuan Survey and Design Group Co.Ltd. Wuhan 430063China National-Local Joint Engineering Research Center of Underwater Tunnelling Technology Wuhan 430063China

Robotic splicing of steel arches is a challenging task that is necessary to realize the grasping and docking of steel arches in a limited *** arches often have a mass of more than 200 kg and length of more than 4 *** to the large volume and mass of steel arches and the high requirements for accurately positioning the splicing,it is difficult for a general manipulator to meet the stiffness *** enhance the structural stiffness of the steel arch splicing manipulator,a single-degree-of-freedom(DOF)closed-loop mechanism was added to the grasping structure of the *** on the basic principle of structural synthesis,a solution model of the single-DOF closed-loop mechanism was developed,and alternative kinematic pairs of the mechanism with different input constraints and output requirements were *** on this model,a design method for a single-DOF closed-loop grasping mechanism and a posture adjustment mechanism for a steel arch was *** with the same dimensional subspace equivalence principle of the graphical-type synthesis method,12 types of steel arch splicing manipulator were *** analyzing the motion/force transmission and structural complexity of the steel arch splicing manipulators,the best scheme was selected.A prototype of the steel arch splicing manipulator was *** software was used to obtain clearly the output trajectory of the end of the *** relative spatial positions of the upper and lower jaws under different working stages were analyzed,demonstrating that the manipulator satisfied the grasping *** a steel arch splicing experiment,the grasping effect,docking accuracy,and splicing efficiency of the manipulator met the design *** steel arch splicing manipulator can replace the manual completion of the steel arch splicing operation,significantly improving the operation efficiency.

关键词： Steel arch splicing manipulator Single-DOF closed-loop mechanism Screw theory Graphical-type synthesis method Mechanism design

来源：评论

学校读者我要写书评

暂无评论

Towards Efficient and Diverse Generative Model for Unconditional Human Motion Synthesis 24

Towards Efficient and Diverse Generative Model for Unconditi...

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Yu, Hua Liu, Weiming Bai, Jiapeng Gui, Xu Hou, Yaqing Ong, Yewsoon Zhang, Qiang Key Laboratory of Social Computing and Cognitive Intelligence Ministry of Education School of Computer Science and Technology Dalian University of Technology Dalian China College of Computer Science Zhejiang University Hangzhou China National and Local Joint Engineering Laboratory of Computer Aided Design School of Software Engineering Dalian University Dalian China College of Computing and Data Science Nanyang Technological University Singapore Singapore Yonsei University Seoul Korea Republic of

ISBN: (纸本)9798400706868

Recent generative methods have revolutionized the way of human motion synthesis, such as Variational Autoencoders (VAEs), Generative Adversarial Networks (GANs), and Denoising Diffusion Probabilistic Models (DMs). These methods have gained significant attention in human motion fields. However, there are still challenges in unconditionally generating highly diverse human motions from a given distribution. To enhance the diversity of synthesized human motions, previous methods usually employ deep neural networks (DNNs) to train a transport map that transforms Gaussian noise distribution into real human motion distribution. According to Figalli's regularity theory, the optimal transport map computed by DNNs frequently exhibits discontinuities. This is due to the inherent limitation of DNNs in representing only continuous maps. Consequently, the generated human motions tend to heavily concentrate on densely populated regions of the data distribution, resulting in mode collapse or mode mixture. To address the issues, we propose an efficient method called MOOT for unconditional human motion synthesis. First, we utilize a reconstruction network based on GRU and transformer to map human motions to latent space. Next, we employ convex optimization to match the noise distribution with the latent space distribution of human motions through the Optimal Transport (OT) map. Then, we combine the extended OT map with the generator of reconstruction network to generate new human motions. Thereby overcoming the issues of mode collapse and mode mixture. MOOT generates a latent code distribution that is well-behaved and highly structured, providing a strong motion prior for various applications in the field of human motion. Through qualitative and quantitative experiments, MOOT achieves state-of-the-art results surpassing the latest methods, validating its superiority in unconditional human motion generation. © 2024 ACM.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

BRG: Bidirectional Regularization Guidance for Unsupervised Domain Adaptation Semantic Segmentation

BRG: Bidirectional Regularization Guidance for Unsupervised ...

引用

International joint Conference on Neural Networks (IJCNN)

作者： Chaoyu Rao Wanshu Fan Xing Yang Xiaopeng Wei Dongsheng Zhou School of Software Engineering National and Local Joint Engineering Laboratory of Computer Aided Design Dalian University Dalian China School of Computer Science and Technology Dalian University of Technology Dalian China

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

Unsupervised Domain Adaptation Semantic Segmentation (UDASS) aims to harness labeled data from the source domain alongside unlabeled data from the target domain to effectively segment target domain images. While self-training has achieved tremendous success in UDASS, most existing methods heavily rely on source data and lack of sufficient learning from target data, resulting in a performance decline. To address this issue, this paper introduces a Bidirectional Regularization Guidance (BRG) method, which combines a Target Perturbation Consistency (TPC) module and a thing-class ImageNet Feature Distance Reweighting (FDR) module to provide effective regularization guidance. Specifically, the TPC module is introduced to perturb the target stream by masking at the input level and adding feature noise at the feature level. This module employs a target perturbation consistency loss to penalize inconsistencies between perturbed student predictions and teacher model predictions, thereby facilitating improved learning of contextual information by the student model from the target images. To further enhance the model’s generalization ability to the target domain, We propose an FDR module that employs a transferability map to calculate a reweighted graph, helping to effectively align source features with ImageNet features by reweighting the feature distances, especially those with lower transferability Through rigorous experimentation in standard UDASS settings—training with synthetic labeled and real unlabeled data—BRG outperforms baseline model on GTAV→Cityscapes and SYNTHIA→Cityscapes.

关键词： Adaptation models Perturbation methods Semantic segmentation Noise Neural networks Object detection Predictive models

来源：评论

学校读者我要写书评

暂无评论

Parallel type based on Si-IGBT and SiC-MOSFET inverter Research on Multifunctional Power Supply 1

Parallel type based on Si-IGBT and SiC-MOSFET inverter Resea...

引用

1st IEEE International Power Electronics and Application Symposium, PEAS 2021

作者： Liu, Shengyu Peng, Zishun Li, Li Zhao, Zhenxing Zhou, Weng Dai, Yuxing Wenzhou University National-Local Joint Engineering Laboratory of Digitalize Electrical Design Technology Wenzhou China Zhejiang CHINT Electrics Co Ltd Electrical Research Institute Yueqing China Hunan Institute of Engineering College of Electrical And Information Engineering Xiangtan China Changsha Hitachi Electric Co Ltd Electrical Research Institute Changsha China

ISBN: (纸本)9781665413596

To improve the problem that the single power device inverter power supply cannot meet the requirements of power quality control under the new situation due to the poor high-frequency performance or high-cost limitation, this paper adopts the harmonic current frequency division control technology combined with the characteristics of Si IGBT and SiC MOSFET devices, and proposes a parallel multi-functional power supply scheme based on Si IGBT and SiC MOSFET inverters to compensate the harmonic current. The high-precision current tracking control is carried out through the fractional-order model predictive control, which improves the current control accuracy of the inverter power supply. The simulation and experimental results show that the scheme is safe, reliable and has various functions, and has a good ability of harmonic control and reactive power compensation. © 2021 IEEE.

关键词： Insulated gate bipolar transistors (IGBT)

来源：评论

学校读者我要写书评

暂无评论

Melatonin regulates lily bulblet development through the LoBPM3- LoRAV module

Ornamental Plant Research

引用

Ornamental Plant Research 2022年第1期2卷 123-133页

作者： Liang Sun Sheng-li Song Yue Yang Hong-mei Sun Key Laboratory of Protected Horticulture of Education Ministry College of HorticultureShenyang Agricultural UniversityShenyang 110161China National and Local Joint Engineering Research Center of Northern Horticultural Facilities Design and Application Technology Shenyang nullChina

Melatonin regulates lily bulblet development;however,the key interactions have not yet been *** this study,we found that adding 20μM melatonin to the growth medium increased the bulblet circumference and the trehalose and starch ***,the expression of numerous genes involved in cell proliferation and carbohydrate accumulation were *** analysis showed that melatonin downregulated the expression of LoRAV and upregulated the expression of LoBPM3 at an early *** and bioinformatics analysis showed that LoRAV is 1056 bp in length and encodes a protein with two structural domains,namely,AP2 and B3;the encoded protein is localized to the nucleus and has no self-activation activity,and this protein interacts with *** expression of LoTCP4 and LoAMY was upregulated,while the expression of LoAGPase,LoCYCB2-1,and LoPCNA was inhibited when LoRAV was transiently ***,the trends in the expression of the genes mentioned above,except LoAGPase and LoAMY,were reversed when LoRAV-OE was treated with ***,the expression of LoRAV,LoTCP4 and LoAMY was inhibited,and the expression of LoCYCB2-1 was upregulated by transient overexpression of LoBPM3,but the expression of LoBPM3 was not regulated by LoRAV *** on these data,we suggest that melatonin promoted bulb development,which was partly associated with regulation of the LoBPM3-LoRAV module;thus,LoRAV negatively regulates cell proliferation of bulblets,and LoBPM3 is located upstream of LoRAV and attenuates its expression.

关键词： expression inhibited Melatonin

来源：评论

学校读者我要写书评

暂无评论

A Fractional Order Model Predictive Control Method for Electric-Hydrogen Hybrid Energy Storage System

A Fractional Order Model Predictive Control Method for Elect...

引用

2021 China Automation Congress, CAC 2021

作者： Xu, Xiaofeng Peng, Zishun Li, Li Zhou, Wen Zhao, Zhenxing Dai, Yuxing National-local Joint Engineering Laboratory For Digitalize Electrical Design Technology Wenzhou University Wenzhou China Zhejiang CHINT Electrics Co. Ltd Yueqing China Changsha Sunye Electric Co. Ltd. Changsha China College of Electrical and Information Engineering Hunan Institute of Engineering Changsha China

ISBN: (纸本)9781665426473

A Fractional order model predictive control (FO-MPC) method for electric-hydrogen combined hybrid energy storage system is proposed in this paper. This method can help the converter of HESS to smooth the fluctuations of DG's output better, improve the control accuracy of the converter output current and improve the stability of DC microgrid. Firstly, the structure of electric-hydrogen combined HESS is illustrated and DC microgrid system is described. Secondly, the concept of fractional calculus is introduced and the predictive model of DC Converter based on fractional order differential equation is established. According to the predictive model, the next current state of bidirectional converter of battery-ultracapacitor and converter of electrolyzer and fuel cell is predicted precisely. Finally, the fractional order model predictive control approach is experimentally verified in Buck/Boost circuit and significantly reduce the current ripple on the DC bus of the microgrid. © 2021 IEEE

关键词： Model predictive control

来源：评论

学校读者我要写书评

暂无评论

BPSO-AHDL-IDS: Binary Particle Swarm Optimization-Based Automated Hybrid Deep Learning Model for Intrusion Detection of Internet of Things

引用

IEEE Transactions on Automation Science and engineering 2025年 22卷 15859-15877页

作者： Lu, Kang-Di Yang, Yao-Wei Zeng, Guo-Qiang Peng, Chen Geng, Guang-Gang Weng, Jian Donghua University College of Information Sciences and Technology Shanghai 201620 China Jinan University College of Cyber Security National Joint Engineering Research Center of Network Security Detection and Protection Technology Guangzhou 510632 China Wenzhou University National-Local Joint Engineering Research Center of Digitalized Electrical Design Technology Wenzhou 325035 China Shanghai University School of Mechatronic Engineering and Automation Shanghai 200444 China

The pervasive adoption of Internet-of-Things (IoT) systems has exposed critical vulnerabilities in cyber-security frameworks due to their decentralized deployment in unattended environments. While deep learning-based intrusion detection systems (IDSs) offer promising solutions, the design of hyper-parameters and neural architectures in existing models imposes prohibitive computational costs and expert dependency. To address these limitations, this work proposes an innovative automated hybrid deep learning method for IDS by employing a binary particle swarm optimization (BPSO) algorithm called BPSO-AHDL-IDS to effectively address the intrusion detection tasks of IoT. In BPSO-AHDL-IDS, the combination of convolutional neural network and recurrent neural network is considered as the hybrid deep learning model to extract features of the IoT dataset for accurate detection of intrusions. First, an efficient binary encoding mechanism is developed to describe the hyper-parameters and neural architectures of the hybrid deep learning model. Then, an efficient BPSO-based evolutionary operation is introduced to evolve the hyper-parameters and neural architectures to discover the optimized hybrid deep learning model. The performance of the proposed BPSO-AHDL-IDS method is testified by employing four datasets gathered from different IoT scenarios. It achieves accuracy of 0.9832, 0.9959, and 0.9897, precision of 0.9700, 0.9788, and 0.9843, recall of 0.9724, 0.9822, and 0.965, and F1-score of 0.9712, 0.9804, and 0.9744 on Bot-IoT, ToN-IoT, and Gas Pipeline datasets, respectively. On SWaT dataset, it achieves a precision of 0.9962, a recall of 0.9969, and an F1-score of 0.9965, respectively. The experimental results show the superiority of the proposed BPSO-AHDL-IDS method to machine learning methods, state-of-the-art manually designed and automated deep learning-based IDSs in terms of accuracy, precision, recall, and F1-score. © 2004-2012 IEEE.

关键词： automated deep learning binary particle swarm optimization hybrid deep learning Internet-of-Things intrusion detection

来源：评论

学校读者我要写书评

暂无评论

CmABF1 and CmCBF4 cooperatively regulate putrescine synthesis to improve cold tolerance of melon seedlings

引用

园艺研究（英文） 2022年第5期9卷 608-621页

作者： Meng Li Xiaoyu Duan Ge Gao Tao Liu Hongyan Qi College of Horticulture Shenyang Agricultural UniversityShenyang 110866LiaoningChina Key Laboratory of Protected Horticulture of Education Ministry and Liaoning Province Shenyang 110866LiaoningChina National and Local Joint Engineering Research Centre of Northern Horticultural Facilities Design and Application Technology(Liaoning)Shenyang 110866LiaoningChina

Low temperatures severely restrict melon seedling ***,the mechanisms by which melon adapts to cold stress are poorly *** decarboxylase(ADC),a key synthetase,catalyzes putrescine biosynthesis in *** this study,we found that CmADC functions as a positive regulator of melon seedling cold *** addition,two transcription factors,abscisic acid-responsive element(ABRE)-binding factor 1(CmABF1)and C-repeat binding factor 4(CmCBF4),directly target CmADC to trigger its ***,virus-induced gene silencing(VIGS)of CmABF1 or CmCBF4 downregulated CmADC abundance,decreased putrescine accumulation,and reduced cold ***,some other CBF and ABF members show at least partial functional redundancy and complementarity with CmABF1 and ***,our work reveals that the ABA,CBF,and polyamine pathways may form a cooperative regulatory network to participate in plant cold stress response.

关键词：

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：