检索结果-内蒙古大学图书馆

3rd International Conference on Optics and Machine Vision, ICOMV 2024

作者： Xie, Qiyang Zhou, Daiying Liu, Zhaozuo School of Information and Communication Engineering University of Electronic Science and Technology of China Chengdu611731 China

ISBN: (纸本)9781510680319

In remote sensing imagery, the detection of fine-grained objects can be seen as a challenging issue, there are 3 conundrums in fine-grained objects detection that cannot be well addressed by the existing deep learning-based methods, the first one is hard to adapt multi-scale objects, the second one and the last is that which have a slow convergence speed and limitations to scarce datasets respectively. To deal with the conundrum we noted before, we propose a deep convolutional neural network (CNN) with a residual structure as the backbone network, to extract deep-level details from the image, which is a mainstream detection method for multi-scale fine-grained objects in complex remote sensing scenarios. Besides, we introduce a multi-scale region generation network to overcome the limitations of fixed receptive field convolution kernels and enable multi-scale object detection. Lastly, we replace fully connected layers in the fully convolutional region classification network with 1×1 convolutional layer to enhance detection efficiency and detection speed. To overcome the limitation of scarce datasets, we conducted experiments on the FAIR1M dataset, which is currently the largest fine-grained object detection dataset in the remote sensing field. Simulation results show that the proposed detection method achieves the highest average precision (35.86%) among all benchmarks and outperforms the classic Faster R-CNN-based method by 3.44%. © 2024 SPIE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Reconfigurable Intelligent Surface for High-Speed Railway mmWave communication System: A Deep Reinforcement Learning Approach 100

Reconfigurable Intelligent Surface for High-Speed Railway mm...

引用

100th IEEE Vehicular technology Conference, VTC 2024-Fall

作者： Talla Nana, Yannick Abel Liu, Gang Tienin, Bole Wilfried Southwest Jiaotong University School of Information Science and Technology Chengdu China University of Electronic Science and Technology of China School of Information and Communication Engineering Chengdu China

ISBN: (纸本)9798331517786

This paper presents research on the use of multiple reconfigurable intelligent surfaces (RIS) in high-speed railway (HSR) tunnel communication systems for signal quality improvement and spectral efficiency (SE) maximization. This is done by placing multiple RISs on opposite sides of the tunnel, which then relay the signal to the onboard mobile relay (MR) of the high-speed train (HST), as it passes through the tunnel. To achieve our objective, a joint beamforming at Basestation (BS) and RISs phase shift SE optimization problem is formulated. The formulated problem is non-convex and hard to solve using existing conventional mathematical optimization techniques. We proposed a special class of deep reinforcement learning (DRL) called rank-based batch prioritization replay deep deterministic policy gradient (RBPR-DDPG) to solve the problem. The performance of the proposed solution is then compared with an adapted version of benchmark uniform random batch replay deep deterministic policy gradient (URBR-DDPG). Simulation results show that the placement of multiple RISs along the HSRtunnel line not only leads to signal quality improvement in the tunnel but also the maximization of the signal SE at the HST mobile relay. © 2024 IEEE.

关键词： Railroads

来源：评论

学校读者我要写书评

暂无评论

UNSUPERVISED FEATURE REPRESENTATION FOR OCEAN-GOING SHIP DETECTION

UNSUPERVISED FEATURE REPRESENTATION FOR OCEAN-GOING SHIP DET...

引用

IET International Radar Conference 2023, IRC 2023

作者： Li, Haoyan Wang, Zhuowei Zhang, Zekai Zhou, Shichao Beijing Information Science & Technology University School of Information and Communication Engineering Beijing China

ISBN: (纸本)9781839539954

Feature representation plays a crucial role in ocean-going ships detection with remote sensing imagery. The robustness of traditional hand-craft features is heavily influenced by prior knowledge, and modern deep neural network models require a large amount of annotated data. In this case, the large field of view characteristic of remote sensing imageries leads to high and challenging annotation costs. Here, we investigate an unsupervised feature representation method, utilizing the spatial context of similar affinity to enhance local descriptors. The refined local descriptors are embedded into a subspace constructed by learned landmarks, aiming to achieve a linear approximation and derive equivalent feature representation. Further, visual cosine similarity is utilized for labelling similar samples in a large-scale data environment. Compared with one-shot object detection methods with classical hand-craft features, the experimental results demonstrate that the proposed method achieves superior detection efficiency with little training instances. © The Institution of engineering & technology 2023.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

An Enhancement Method with Autoencoder for Deep Learning Based Hybrid Beamforming 4

An Enhancement Method with Autoencoder for Deep Learning Bas...

引用

4th International Conference on Neural Networks, information and communication engineering, NNICE 2024

作者： Xie, Junhan Luo, Chunbo Liu, Zijian Luo, Yang University of Electronic Science and Technology of China School of Information and Communication Engineering Chengdu611731 China

ISBN: (纸本)9798350394375

This paper proposes a universal performance enhancement method for the existing hybrid beamforming structures based on deep learning (DL). The input of the existing DL-based hybrid beamforming (DL-BF) models only comprises the real and imaginary parts of the complex channel. The information hidden in the two parts tends to be ignored, leading to performance degradation. Based on the fact, the proposed method utilizes an autoencoder consist of the fully-connected layers to correlate the information between the real and imaginary parts of the complex channel matrix. The output feature from the intermediate layer of the autoencoder is extracted as enhancement information. This information is then concatenated with the original real and imaginary components of the complex channel matrix, creating an enhanced dataset for the existing DL-BF models to improve the performance. Simulation results demonstrate that our proposed method can generally improve the spectral efficiency of the state-of-art DL-BF models. © 2024 IEEE.

关键词： Spectrum efficiency

来源：评论

学校读者我要写书评

暂无评论

Green Peafowl Detection System for Paddy Field Value Creation in Phayao Area 9

Green Peafowl Detection System for Paddy Field Value Creatio...

引用

Joint 9th International Conference on Digital Arts, Media and technology with 7th ECTI Northern Section Conference on Electrical, Electronics, Computer and Telecommunications engineering, ECTI DAMT and NCON 2024

作者： Chaiwongsai, Jirabhorn Udomsripaiboon, Thana Srisungsittisunti, Bowonsak Jeefoo, Phaisarn Amorndechaphon, Damrong Ratchakom, Mathaya University of Phayao School of Information and Communication Technology Phayao Thailand University of Phayao School of Engineering Phayao Thailand

ISBN: (纸本)9798350318241

Green peafowl is classified as protected wildlife of the Wildlife Preservation and Protection Act Thailand and the International Union for Conservation of Nature (IUCN). The green peafowl can be found in Phayao area considered as the stronghold of the green peafowl. However, farmers, who have paddy fields located nearby the forest patch where the peafowl lives, currently suffer with the peafowl eating their agricultural products. This results in lower yields and lower incomes. To support sustainable coexistence between communities and peafowl, and promote a circular economy in the area, this paper proposes green peafowl detection system for paddy field value creation in Phayao area. The system is available to detect the green peafowls using camera modules that can be stored and manage data sending schedule online. The system is designed to be able to set and view videos of the green peafowl behavior via web application. Therefore, it can be used as a tool building trust between farmers and customers efficiently. © 2024 IEEE.

关键词： Cameras

来源：评论

学校读者我要写书评

暂无评论

High-Precision Attitude Measurement Based on Differential MEMS Sensor Array Structure and Improved ESKF Algorithm 5

High-Precision Attitude Measurement Based on Differential ME...

引用

5th International Conference on Sensing, Measurement and Data Analytics in the Era of Artificial Intelligence, ICSMD 2024

作者： Zhu, Weiye Zhang, Jinquan Wang, Maolin University of Electronic Science and Technology of China School of Information and Communication Engineering Chengdu611731 China

ISBN: (纸本)9798331529192

High-precision attitude measurement is widely used in aerospace, robotics, medical imaging and other fields. The precision of attitude measurement mainly depends on MEMS inertial sensor array technology and advanced attitude solution algorithms, but the existing sensor array design and attitude solution algorithms can not effectively deal with the measurement noise. To solve the above problems, an attitude measurement system based on differential sensor array structure is designed in this paper, and an Improved Error State Kalman Filter (IESKF) algorithm based on residual extraction and analysis technology is proposed. The experimental results show that both the differential sensor array and IESKF algorithm can effectively improve the accuracy of attitude measurement. The angular velocity error of the differential array is improved by about 79.4% compared with that of a single sensor, and that of a group of differential structure sensors is improved by about 31.4% compared with that of a dual sensor. The quaternion output accuracy of IESKF algorithm is better than that of the general error state Kalman filtering algorithm. © 2024 IEEE.

关键词： Kalman filters

来源：评论

学校读者我要写书评

暂无评论

Combined Classification of Hyperspectral and LiDAR Data based on Dual-Channel Cross-Transformer 5

Combined Classification of Hyperspectral and LiDAR Data base...

引用

5th International Conference on Video, Signal and Image Processing, VSIP 2023

作者： Zhou, Binbin Wang, Qingyan Zhang, Junping Wang, Yujing School Of: Measurement-Control And Communication Engineering Harbin University Of Science And Technology Harbin150080 China School Of: Electronics And Information Engineering Harbin Institute Of Technology Harbin150001 China

ISBN: (纸本)9798400709272

In the face of complex scenes, single-modal dominant classification tasks encounter limitations in performance due to insufficient information. On the other hand, joint classification of multimodal remote sensing data faces challenges such as data sample differences and lack of correlation in physical features between modalities, which can impact classification performance. To fully integrate the heterogeneous information of multimodal data and improve classification performance, we propose a dual-channel cross-transformer feature fusion extraction network. The framework leverages self-attention mechanisms to aggregate features within each modality, and the feature fusion module based on cross-modal attention fully considers the complementary information between modalities. Classification tasks are performed using the fused spatial-spectral features obtained from the joint representation of modalities. Extensive experiments conducted on the Houston and MUUFL datasets demonstrate the effectiveness of the proposed model compared to existing methods. © 2023 ACM.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Low Latency Soft Fiberoptic Choledochoscope Robot control System

引用

Journal of Shanghai Jiaotong university(Science) 2025年第1期30卷 43-52页

作者： ZHOU Hanwei ZHU Xinping MA Youwei WANG Kundong School of Electronic Information and Electrical Engineering Shanghai Jiao Tong University200240ShanghaiChina Shanghai Aerospace Control Technology Institute Shanghai 201109China

Soft fiberoptic choledochoscope is an important tool for the diagnose and surgical treatment of biliary ***,the traditional soft fiberoptic choledochoscope is hard to operate,due to the low position *** on the conventional soft fiberoptic choledochoscope,an electrical soft fiberoptic choledochoscope robot with a low latency was *** order to improve the controllability of the conventional choledochoscope,the wire traction mechanism and the rotation mechanism are used to bend and rotate the scope,so as to control its movement *** dead band compensation model and control algorithm of the wire traction mechanism are developed to improve the accuracy of the orientation *** human-computer interaction system and complex motion control system are developed based on ARM embedded system and *** to the high-speed synchronization channel between FPGA and peripheral,the design of low latency whole-procedure surgical mode was established and *** with a micro image sensor,real-time video back transmission was *** performance of the robot prototype was verified by animal experiment in vivo on a *** robot has an extremely low operating latency of no more than 0.402 ms,and a high bending positioning accuracy of±1.43°error margin within 99.7%confidence interval,which guarantees the safety of biliary surgery.

关键词： soft fiberoptic choledochoscope electrical control low latency in vivo animal experiment

来源：评论

学校读者我要写书评

暂无评论

Waveform Design for Wireless Powered and Backscatter Hybrid communication System

引用

Journal of communications and information Networks 2024年第4期9卷 309-319页

作者： Yuxin Ding Bin Rao Jie Hu School of Information and Communication Engineering University of Electronic Science and Technology of ChinaChengdu 611731China

For the waveform design and research of wireless powered and backscatter hybrid communication system,it is crucial to balance energy harvesting with ensuring communication rate *** the communication needs of traditional users in the hybrid communication system is particularly *** this paper,we study the model of wireless powered and backscatter hybrid communication system,while also taking the traditional orthogonal frequency division multiplexing(OFDM)user system into *** jointly design the transmitting signal waveform and the backscatter signal waveform to ensure the communication performance of the traditional user while simultaneously enhancing the communication and energy transmission performance of the hybrid communication *** maximize the signal-to-noise ratio(SNR)at the receiver by jointly optimizing the amplitude and phase of the transmitted signal waveform from the radio frequency(RF)source and the reflection coefficient of the backscatter device(BD).Furthermore,we use geometric programming methods to solve the *** simulation results confirm the effectiveness of the proposed scheme.

关键词： backscatter communication wireless powered communication OFDM waveform optimization design

来源：评论

学校读者我要写书评

暂无评论

Quantum walk search algorithm for multi-objective searching with iteration auto-controlling on hypercube

引用

Chinese Physics B 2022年第4期31卷 157-162页

作者： Yao-Yao Jiang Peng-Cheng Chu Wen-Bin Zhang Hong-Yang Ma School of Science Qingdao University of TechnologyQingdao 266033China School of Information and Control Engineering Qingdao University of TechnologyQingdao 266033China

Shenvi et *** proposed a quantum algorithm based on quantum walking called Shenvi-Kempe-Whaley(SKW)algorithm,but this search algorithm can only search one target state and use a specific search target state ***,when there are more than two target nodes in the search space,the algorithm has certain *** though a multiobjective SKW search algorithm was proposed later,when the number of target nodes is more than two,the SKW search algorithm cannot be mapped to the same quotient *** addition,the calculation of the optimal target state depends on the number of target states *** previous studies,quantum computing and testing algorithms were used to solve this *** these solutions require more Oracle calls and cannot get a high accuracy ***,to solve the above problems,we improve the multi-target quantum walk search algorithm,and construct a controllable quantum walk search algorithm under the condition of unknown number of target *** dividing the Hilbert space into multiple subspaces,the accuracy of the search algorithm is improved from p_(c)=(1/2)-O(1/n)to p_(c)=1-O(1/n).And by adding detection gate phase,the algorithm can stop when the amplitude of the target state becomes the maximum for the first time,and the algorithm can always maintain the optimal number of iterations,so as to reduce the number of unnecessary iterations in the algorithm process and make the number of iterations reach t_(f)=(π/2)(?).

关键词： multi-objective quantum walk search algorithm accurate probability

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：