检索结果-内蒙古大学图书馆

SSRN 2024年

作者： Zheng, Wenhao Xiong, Bangshu Yi, Hui Au, Qiaofeng Chen, Jiujiu Key Laboratory of Jiangxi Province for Image Processing and Pattern Recognition Nanchang Hangkong University Nanchang330063 China China Helicopter Research and Development Institute China

Detecting blade tip point light sources based on airborne computer vision is a critical step in measuring blade tip distance for coaxial unmanned helicopters. However, detecting blade tip point light sources quickly and highly accurately in a helicopter airborne environment is a challenge due to the extremely tiny imaging size of the blade tip point light sources. To this end, we propose a FRE-YOLO for detecting blade tip point light sources based on the YOLO network structure. In this work, we focus on designing a Feature Refinement Extraction network (FRE), which can greatly enrich the feature information of small targets during Backbone downsampling. Secondly, we also design a new paradigm Smooth Gaussian Distance (SmoothGD) which measures the degree of target bounding box matching. SmoothGD is mainly applied to regression loss calculation and label assigner, which can effectively improve the small target detection accuracy. Meanwhile, we construct the Blade Tip Point Light (BTPL) dataset for the network training. In addition, in order to evaluate the effectiveness of the proposed method more fairly, this paper conducts the same experiments on BTPL and the small target dataset VisDrone on MMYOLO. A large amount of experimental data shows that the proposed method has a positive impact on the detection of tiny point sources at the tip of the blade. It is also at the advanced level in the field of generalized small object detection. © 2024, The Authors. All rights reserved.

关键词： Extraction

来源：评论

学校读者我要写书评

暂无评论

HyRSM++: Hybrid relation guided temporal set matching for few-shot action recognition

引用

Pattern Recognition 2024年 147卷

作者： Wang, Xiang Zhang, Shiwei Qing, Zhiwu Zuo, Zhengrong Gao, Changxin Jin, Rong Sang, Nong Key Laboratory of Ministry of Education for Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology China Alibaba Group China Meta AI United States

Few-shot action recognition is a challenging but practical problem aiming to learn a model that can be easily adapted to identify new action categories with only a few labeled samples. However, existing attempts still suffer from two drawbacks: (i) learning individual features without considering the entire task may result in limited representation capability, and (ii) existing alignment strategies are sensitive to noises and misaligned instances. To handle the two limitations, we propose a novel Hybrid Relation guided temporal Set Matching (HyRSM++) approach for few-shot action recognition. The core idea of HyRSM++ is to integrate all videos within the task to learn discriminative representations and involve a robust matching technique. To be specific, HyRSM++ consists of two key components, a hybrid relation module and a temporal set matching metric. Given the basic representations from the feature extractor, the hybrid relation module is introduced to fully exploit associated relations within and cross videos in an episodic task and thus can learn task-specific embeddings. Subsequently, in the temporal set matching metric, we carry out the distance measure between query and support videos from a set matching perspective and design a bidirectional Mean Hausdorff Metric to improve the resilience to misaligned instances. Furthermore, we extend the proposed HyRSM++ to deal with the more challenging semi-supervised few-shot action recognition and unsupervised few-shot action recognition tasks. Experimental results on multiple benchmarks demonstrate that our method consistently outperforms existing methods and achieves state-of-the-art performance under various few-shot settings. The source code is available at https://***/alibaba-mmai-research/HyRSMPlusPlus. © 2023 Elsevier Ltd

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

A Recipe for Scaling up Text-to-Video Generation with Text-free Videos

A Recipe for Scaling up Text-to-Video Generation with Text-f...

引用

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Xiang Wang Shiwei Zhang Hangjie Yuan Zhiwu Qing Biao Gong Yingya Zhang Yujun Shen Changxin Gao Nong Sang Key Laboratory of Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology Alibaba Group Zhejiang University Ant Group

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

Diffusion-based text-to-video generation has witnessed impressive progress in the past year yet still falls behind text-to-image generation. One of the key reasons is the limited scale of publicly available data (e.g., 10M video-text pairs in WebVid10m vs. 5B image-text pairs in LAION), considering the high cost of video captioning. Instead, it could be far easier to collect unlabeled clips from video platforms like YouTube. Motivated by this, we come up with a novel text-to-video generation framework, termed TF-T2V, which can directly learn with text-free videos. The rationale behind is to separate the process of text decoding from that of temporal modeling. To this end, we employ a content branch and a motion branch, which are jointly optimized with weights shared. Following such a pipeline, we study the effect of doubling the scale of training set (i.e., video-only WebVid10M) with some randomly collected text-free videos and are encouraged to observe the performance improvement (FID from 9.67 to 8.19 and FVD from 484 to 441), demonstrating the scalability of our approach. We also find that our model could enjoy sustainable performance gain (FID from 8.19 to 7.64 and FVD from 441 to 366) after reintroducing some text labels for training. Finally, we validate the effectiveness and generalizability of our ideology on both native text-to-video generation and compositional video synthesis paradigms. Code and models will be publicly available at here.

关键词： Training Video on demand Scalability Pipelines Text to image Performance gain Market research

来源：评论

学校读者我要写书评

暂无评论

Reward Shaping-based Double Deep Q-networks for Unmanned Surface Vessel Navigation and Obstacle Avoidance 48

Reward Shaping-based Double Deep Q-networks for Unmanned Sur...

引用

48th Annual Conference of the IEEE Industrial Electronics Society, IECON 2022

作者： Gan, Zihan Zheng, Jinghong Jiang, Zhenyu Lu, Renzhi Huazhong University of Science and Technology Key Laboratory of Image Processing and Intelligent Control School of Artificial Intelligence and Automation Wuhan430074 China Wuhan University of Science and Technology Hubei Key Laboratory of Mechanical Transmission and Manufacturing Engineering Wuhan430081 China

ISBN: (数字)9781665480253

ISBN: (纸本)9781665480253

In this paper, a method for navigation and obstacle avoidance of unmanned surface vessel (USV) based on reinforcement learning and reward shaping is proposed. This approach uses double deep Q networks (DDQN) to make decisions based on the continuous states observed from sensors in USV. In addition, a new reward function is designed based on prior knowledge to accelerate the convergence of the algorithm and improve the performance. For training the neural networks, a simulation platform is developed, in which a 3 degree of freedom mathematical model describes USV dynamic system and two-dimension actions are required to control USV. Simulation results on the platform demonstrate the DDQN hoists USV's capabilities of navigation and obstacle avoidance, and reward shaping technique improves the speed of convergence. © 2022 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Set pair three-way overlapping community discovery algorithm for weighted social internet of things

引用

Digital communications and Networks 2023年第1期9卷 3-13页

作者： Chunying Zhang Jing Ren Lu Liu Shouyue Liu Xiaoqi Li Liya Wang College of Science North China University of Science and TechnologyTangshanHebeiChina Hebei Engineering Research Center for the Intelligentization of Iron Ore Optimization and Ironmaking Raw Materials Preparation Processes North China University of Science and TechnologyTangshanHebeiChina Hebei Key Laboratory of Data Science and Application North China University of Science and TechnologyTangshanHebeiChina The Key Laboratory of Engineering Computing in Tangshan City North China University of Science and TechnologyTangshanHebeiChina Tangshan Intelligent Industry and Image Processing Technology Innovation Center North China University of Science and TechnologyTangshanHebeiChina

There are many problems in Social Internet of Things(IoTs),such as complex topology information,different degree of association between nodes and overlapping *** idea of set pair information grain computing and clustering is introduced to solve the above problems so as to accurately describe the similarity between nodes and fully explore the multi-community structure.A Set Pair Three-Way Overlapping Community Discovery Algorithm for Weighted Social Internet of Things(WSIoT-SPTOCD)is *** the local network structure,which fully considers the topological information between nodes,the set pair connection degree is used to analyze the identity,difference and reverse of neighbor *** similarity degree of different neighbor nodes is defined from network edge weight and node degree,and the similarity measurement method of set pair between nodes based on the local information structure is *** to the number of nodes'neighbors and the connection degree of adjacent edges,the clustering intensity of nodes is defined,and an improved algorithm for initial value selection of k-means is *** nodes are allocated according to the set pair similarity between nodes and different ***-way community structures composed of a positive domain,boundary domain and negative domain are generated ***,the overlapping node set is generated according to the calculation results of community node ***,experiments are carried out on artificial networks and real *** results show that WSIoT-SPTOCD performs well in terms of standardized mutual information,overlapping community modularity and F1.

关键词： Social internet of things Set pair analysis K-means Local information structure Overlapping community

来源：评论

学校读者我要写书评

暂无评论

Multi-objective Optimal Operation of Centralized Battery Swap Charging System with Photovoltaic

引用

Journal of Modern Power Systems and Clean Energy 2022年第1期10卷 149-162页

作者： Yuanzheng Li Yihan Cai Tianyang Zhao Yun Liu Jian Wang Lei Wu Yong Zhao School of Artificial Intelligence and Automation Ministry of Education Key Laboratory of Image Processing and Intelligence ControlHuazhong University of Science and TechnologyWuhanChina School of Electrical and Electronic Engineering Nanyang Technological UniversitySingaporeSingapore College of Mechatronics and Control Engineering Shenzhen UniversityShenzhenChina Department of Electrical&Computer Engineering Stevens Institute of TechnologyHobokenUSA

Electric vehicles(EVs)are widely deployed throughout the world,and photovoltaic(PV)charging stations have emerged for satisfying the charging demands of EV *** paper proposes a multi-objective optimal operation method for the centralized battery swap charging system(CBSCS),in order to enhance the economic efficiency while reducing its adverse effects on power *** proposed method involves a multi-objective optimization scheduling model,which minimizes the total operation cost and smoothes load fluctuations,***,we modify a recently proposed multi-objective optimization algorithm of non-sorting genetic algorithm III(NSGA-III)for solving this scheduling ***,simulation studies verify the effectiveness of the proposed multi-objective operation method.

关键词： Multi-objective optimization electric vehicle battery swap charging system scheduling photovoltaic

来源：评论

学校读者我要写书评

暂无评论

Stability Enhancement of Inverter-Current Feedback Grid-Following Inverters Using a Passivity-Based Voltage Controller

引用

IEEE Transactions on Energy Conversion 2025年

作者： Huang, Yukai Xu, Jinbang Ye, Jie Huang, Songtao Ren, Ziao Shen, Anwen Hu, Yihua Huazhong University of Science and Technology National Key Laboratory of Science and Technology on Multispectral Information Processing Key Laboratory of Image Processing and Intelligent Control School of Artificial Intelligence and Automation Wuhan430074 China King's College London Department of Engineering LondonWC2R 2LS United Kingdom

Grid-following inverters remain the primary interface for modern renewable energy systems due to their strong power output performance. As the global demand for clean and efficient energy conversion increases, optimizing the stability of inverter systems is crucial for integrating renewable sources into the power grid. In certain energy conversion scenarios, current sensors are placed on the inverter side due to equipment constraints and space limitations, which can impact the stability of the inverter. Although the stability of inverter-side current feedback (ICF) configurations has been studied, methods to further enhance their stability remain underexplored. In this paper, the stability of ICF grid-following inverters is analyzed using a passivity-based design approach, focusing on a point of common coupling voltage controller. The proposed controller is thoroughly analyzed, and a detailed design strategy is presented to ensure stability under varying grid impedance conditions. Experimental results validate the theoretical analysis, demonstrating the effectiveness of the proposed design in enhancing the stability of ICF grid-following inverters. These advancements contribute to improving energy conversion efficiency and reliability. © 2025 IEEE.

关键词： Energy conversion efficiency

来源：评论

学校读者我要写书评

暂无评论

FoldExplorer: Fast and Accurate Protein Structure Search with Sequence-Enhanced Graph Embedding

arXiv

引用

arXiv 2023年

作者： Liu, Yuan Shen, Hong-Bin Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Key Laboratory of System Control and Information Processing Ministry of Education of China Shanghai China

The advent of highly accurate protein structure prediction methods has fueled an exponential expansion of the protein structure database. Consequently, there is a rising demand for rapid and precise structural homolog search. Traditional alignment-based methods are dedicated to precise comparisons between pairs, exhibiting high accuracy. However, their sluggish processing speed is no longer adequate for managing the current massive volume of data. In response to this challenge, we propose a novel deep-learning approach FoldExplorer. It harnesses the powerful capabilities of graph attention neural networks and protein large language models for protein structures and sequences data processing to generate embeddings for protein structures. The structural embeddings can be used for fast and accurate protein search. And the embeddings also provide insights into the protein space. FoldExplorer demonstrates a substantial performance improvement of 5% to 8% over the current state-of-the-art algorithm on the benchmark datasets. Meanwhile, FoldExplorer does not compromise on search speed and excels particularly in searching on a large-scale dataset. Copyright © 2023, The Authors. All rights reserved.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Dual-mode predictive control of a rotor suspension system

引用

Science China(Information Sciences) 2020年第1期63卷 178-189页

作者： Yue WU Gui-Ping REN Hai-Tao ZHANG Key Laboratory of Image Processing and Intelligent Control School of Artificial Intelligence and AutomationHuazhong University of Science and Technology State Key Laboratory of Digital Manufacturing Equipments and Technology

Rotor active magnetic bearing(rotor-AMB) systems are frequently used to alleviate vibrations for various applications such as in national defense, manufacturing industries, IC production, and aerospace engineering. One obstacle to improve machining efficiency and quality is the open-loop instability of rotorAMB systems during the machining process. We built a closed-loop processing platform using a spindle rotor installed with AMBs and thereby developed a rotor-AMB suspension system embedded with a dual-mode predictive controller(DMPC). The performance of the system is thus substantially improved. In the proposed DMPC, both model-based prediction and receding horizon optimization are utilized to guarantee the closedloop stability of the rotor-AMB suspension systems with input constraints. Finally, the effectiveness and superiority of the proposed method are examined through substantial levitation experiments on a machining platform with installed AMBs.

关键词： dual-mode predictive control system identification optimization control model predictive control suspension control

来源：评论

学校读者我要写书评

暂无评论

A Lightweight Chinese Summarization Generation Method Based on Knowledge Distillation 4

A Lightweight Chinese Summarization Generation Method Based ...

引用

4th International Conference on Information Science, Parallel and Distributed Systems, ISPDS 2023

作者： Liu, Shiyue Xiao, Yannan Wang, Xueyong Huang, Wenming Deng, Zhenrong School of Computer and Information Security Guilin University of Electronic Technology Guangxi Key Laboratory of Image and Graphic Intelligent Processing Guilin China School of Computer and Information Security Guilin University of Electronic Technology Guilin China Guilin Xintong Technology Co. Guilin China

ISBN: (纸本)9798350337181

In recent years, large-scale pretrained natural language processing models such as BERT, GPT3 have achieved good results in processing tasks. However, in daily applications, these large-scale language models usually exist large model size or long running time, and the model portability and application are not very convenient. To settle this problem, we come up with a lightweight summarization generation method based on knowledge distillation (LW-BERT-KD, abbreviated as LBK model) to prove that complex neural network knowledge (taking BERT model as an example) can be extracted to lightweight language processing models (taking BiLSTM model as an example), thus achieving the purpose of complex knowledge extraction and model compression. In this paper, the BERT model is regarded as the teacher model, and the BiLSTM is regarded as the student model. The knowledge distillation technology is the transfer of knowledge from teacher models to student models, so that the generated model is effective, lightweight, and easy to port. On the Chinese abstract generation dataset, the training effect of this model is significantly better than the basic Transformer baseline. The experimental results show that if the training parameters are increased by 100 times, the model achieves comparable results with complex preprocessing language models. © 2023 IEEE.

关键词： Distillation

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：