检索结果-内蒙古大学图书馆

IEEE Transactions on Artificial Intelligence 2025年第5期6卷 1345-1359页

作者： Hui, Xuemeng Liu, Zhunga Liu, Jiaxiang Zhang, Zuowei Wang, Longfei Northwestern Polytechnical University Key Laboratory of Information Fusion Technology of Ministry of Education Shaanxi Province Xi’an710072 China

Zero-shot learning (ZSL) aims to recognize unseen class image objects using manually defined semantic knowledge corresponding to both seen and unseen images. The key of ZSL lies in building the interaction between precise image data and fuzzy semantic knowledge. The fuzziness is attributed to the difficulty in quantifying human knowledge. However, the existing ZSL methods ignore the inherent fuzziness of semantic knowledge and treat it as precise data during building the visual–semantic interaction. This is not good for transferring semantic knowledge from seen classes to unseen classes. In order to solve this problem, we propose a visual–semantic fuzzy interaction network (VSFIN) for ZSL. VSFIN utilize an effective encoder–decoder structure, including a semantic prototype encoder (SPE) and visual feature decoder (VFD). The SPE and VFD enable the visual features to interact with semantic knowledge via cross-attention. To achieve visual–semantic fuzzy interaction in SPE and VFD, we introduce the concept of membership function in fuzzy set theory and design a membership loss function. This loss function allows for a certain degree of imprecision in visual–semantic interaction, thereby enabling VSFIN to becomingly utilize the given semantic knowledge. Moreover, we introduce the concept of rank sum test and propose a distribution alignment loss to alleviate the bias towards seen classes. Extensive experiments on three widely used benchmarks have demonstrated that VSFIN outperforms current state-of-the-art methods under both conventional ZSL (CZSL) and generalized ZSL (GZSL) settings. © 2020 IEEE.

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

Multimodal LIBS-FLIPA fusion with frame segmentation for robust plastic classification via advanced LIPA processing

引用

Optics Letters 2025年第9期50卷 3038-3041页

作者： Meng, Huanqing Gao, Wenhan Ye, Yanpeng Liu, Yuzhu State Key Laboratory Cultivation Base of Atmospheric Optoelectronic Detection and Information Fusion Nanjing University of Information Science & Technology Nanjing210044 China Nanjing University of Information Science & Technology Nanjing210044 China

The global increase in plastic waste, exceeding 400 million tons annually, underscores the urgent need for efficient plastic sorting and recycling. Laser-induced breakdown spectroscopy (LIBS) shows potential in this area, but its practical application is limited by challenges such as plasma fluctuations and low robustness. To address these limitations, we introduce laser-induced plasma acoustic (LIPA) signals and propose the frame-segmentation LIPA (FLIPA) algorithm to enhance LIBS analysis. This innovative algorithm reduces the number of variables in LIPA by 99% while optimizing computational efficiency and classification accuracy. Additionally, a multimodal fusion technique, LIBS-FLIPA, is developed to integrate LIBS and FLIPA at the feature level. The results indicate that LIBS-FLIPA significantly improves classification accuracy, robustness, and generalization, effectively mitigating overfitting risks. This study provides novel, to the best of our knowledge, solutions to challenges in LIBS analysis and proposes an innovative approach for robust plastic sorting, advancing the methodologies of LIBS research. © 2025 Optica Publishing Group. All rights, including for text and data mining (TDM), Artificial Intelligence (AI) training, and similar technologies, are reserved.

关键词： Laser produced plasmas

来源：评论

学校读者我要写书评

暂无评论

On-orbit attitude planning for Earth observation imaging with star-based geometric calibration

引用

Science China(Technological Sciences) 2025年第2期68卷 243-256页

作者： Di HUANG Zhifu WANG Jianya GONG Haoran JI Institute of Aerospace Science and Technology School of Remote Sensing and Information EngineeringWuhan UniversityWuhan 430072China State Key Laboratory of Information Engineering in Surveying Mapping and Remote SensingSchool of Remote Sensing and Information EngineeringWuhan UniversityWuhan 430072China National Key Laboratory of Air-based Information Perception and Fusion Luoyang 471009China Intelligent Game and Decision Laboratory Beijing 100071China

To improve the quality of remote sensing images,a novel on-orbit attitude planning method for Earth observation imaging with star-based geometric calibration is presented in this *** traditional imaging processes,the proposed method includes both pre-calibration and post-calibration stages to enhance the accuracy of satellite geometric positioning.A multiple-constraint attitude planning algorithm was developed to ensure star-based calibration,wherein the satellite camera and star sensors simultaneously capture the images of *** integration of cameras and multiple star sensors enables effective attitude *** the Earth imaging stage,a fast on-orbit attitude planning algorithm was developed to determine the satellite attitude and the optimal imaging times for ground *** with existing methods,the fast on-orbit attitude planning algorithm can significantly reduce computational time and resource consumption via initial value selection and *** demonstrate the effectiveness of the proposed attitude planning method,which was successfully applied to Wuhan 1 satellite.

关键词： attitude planning Earth observation imaging geometric calibration attitude constraints

来源：评论

学校读者我要写书评

暂无评论

Non-Line-of-Sight Multipath Classification Method for BDS Using Convolutional Sparse Autoencoder with LSTM

引用

Tsinghua Science and technology 2025年第1期30卷 68-86页

作者： Yahang Qin Zhenni Li Shengli Xie Bo Li Ming Liu Victor Kuzin School of Automation Guangdong University of Technologyand also with Guangdong-Hong Kong-Macao Joint Laboratory for Smart Discrete ManufacturingGuangzhou 510006China School of Automation Guangdong University of TechnologyGuangdong Key Laboratory of IoT Information TechnologyGuangzhou 510006China 111 Center for Intelligent Batch Manufacturing Based on IoT Technology Key Laboratory of Intelligent Information Processing and System Integration of IoTMinistry of EducationGuangzhou 510006China Department of Electronic and Computer Engineering Hong Kong University of Science and TechnologHong Kong 999077China Academician of Russian Engineering Academy Moscow Moscow 125009Russia

Multipath signal recognition is crucial to the ability to provide high-precision absolute-position services by the BeiDou Navigation Satellite System(BDS).However,most existing approaches to this issue involve supervised machine learning(ML)methods,and it is difficult to move to unsupervised multipath signal recognition because of the limitations in signal *** by an autoencoder with powerful unsupervised feature extraction,we propose a new deep learning(DL)model for BDS signal recognition that places a long short-term memory(LSTM)module in series with a convolutional sparse autoencoder to create a new autoencoder ***,we propose to capture the temporal correlations in long-duration BeiDou satellite time-series signals by using the LSTM module to mine the temporal change patterns in the time ***,we develop a convolutional sparse autoencoder method that learns a compressed representation of the input data,which then enables downscaled and unsupervised feature extraction from long-duration BeiDou satellite series ***,we add an l_(1/2) regularizer to the objective function of our DL model to remove redundant neurons from the neural network while ensuring recognition *** tested our proposed approach on a real urban canyon dataset,and the results demonstrated that our algorithm could achieve better classification performance than two ML-based methods(e.g.,11%better than a support vector machine)and two existing DL-based methods(e.g.,7.26%better than convolutional neural networks).

关键词： convolutional sparse autoencoder BeiDou Navigation Satellite System(BDS) long short-term memory(LSTM) multipath classification

来源：评论

学校读者我要写书评

暂无评论

An Efficient Coherent Integration Method for Highly Maneuvering Target Detection by Exploiting Range-Azimuth Coupling Relationships Between Subsegments

引用

IEEE Sensors Journal 2025年第9期25卷 15456-15476页

作者： Wan, Jun Jiang, Xinyu Liu, Chengyuan Li, Dong Xie, Xiaoran Sang, Qinglin Chen, Zhanye Chongqing University School of Microelectronics and Communication Engineering Chongqing Key Laboratory of Space Information Network and Intelligent Information Fusion Chongqing400044 China China Academy of Electronics and Information Technology Beijing100041 China State Key Laboratory of Millimeter Waves Institute of Electromagnetic Space Southeast University Nanjing210096 China

Arbitrary-order range migrations (RMs) and Doppler frequency migrations (DFMs) caused by complicated motions of highly maneuvering targets (HMTs) pose challenges for coherent integration (CI) in radar remote sensing and detection applications. Complicated Doppler frequency ambiguity improves the challenges of CI for HMTs. The optimal generalized Radon-Fourier transform (GRFT) suffers from a large computational cost, and nonlinear transform-based methods are affected by severe performance loss. To tackle these difficulties, an efficient segmented CI method for HMT detection via the range-azimuth coupling relationship between subsegments is proposed. First, inspired by the advantage of subaperture processing, a segmentation operation is introduced to remove intrasegment RMs and DFMs. Second, a high-order compensation function is presented to remove the high-order intersegment RMs and DFMs. Third, by exploiting the range-azimuth coupling relationship between subsegments, a residual migration compensation function is used to correct the residual intersegment RM. Thus, a well result of HMT's CI is achieved. The proposed method handles arbitrary-order RMs and DFMs, complex Doppler ambiguity, and ambiguous speed sidelobes (BSSLs) in low signal-to-noise ratio (SNR) circumstances and multitarget cases due to its linear transform property. In comparison to the optimal GRFT, the proposed approach substantially lowers computational cost while showing similar performance. Then, the presented approach can be implemented by parallel processing. It is advantageous to enhance real-time processing efficiency. Several real radar data are analyzed to confirm the effectiveness of the developed method. © 2001-2012 IEEE.

关键词： Signal to noise ratio

来源：评论

学校读者我要写书评

暂无评论

Unsupervised domain adaptation via causal-contrastive learning

引用

Journal of Supercomputing 2025年第5期81卷 1-23页

作者： Wei, Xing Jiang, Wenhao Yang, Fan Zhao, Chong Lu, Yang Zhang, Benhong Bi, Xiang School of Computer and Information Hefei University of Technology Hefei230601 China Anhui Mine IOT and Monitoring Technology Key Laboratory Hefei230022 China Engineering Quality Education Center of Undergraduate School Hefei University of Technology Hefei230601 China

Unsupervised domain adaptation (UDA) aims to reduce the domain differences between source and target domains by mapping their data to a shared feature space, thereby learning domain-invariant features. The aim of this study is to address the challenges faced by contrastive learning-based UDA methods when dealing with domain discrepancies, particularly the spurious correlations introduced by confounding factors caused by data augmentation. In recent years, contrastive learning has gained attention for its powerful representation learning capabilities, as it can pull similar samples from the source and target domains closer together while separating different classes of negative samples. This process helps alleviate domain differences and enhances the model’s generalization ability. However, mainstream UDA methods based on contrastive learning often introduce confounding factors due to the randomness of data augmentation, leading the model to learn incorrect spurious associations, especially when the target domain contains counterfactual data from the source domain. As the amount of counterfactual data increases, this bias and accuracy loss can significantly exacerbate and are difficult to eliminate through non-causal methods. To address this, this paper proposes causal invariance contrastive adaptation (CICA), a causal-contrastive learning-based unsupervised domain adaptation model for image classification. The model inputs labeled source domain samples and unlabeled target domain samples into a feature generator after data augmentation, and quantifies the degree of confusion between the generated features based on a backdoor criterion. We effectively separate domain-invariant features from spurious features using adversarial training, thereby reducing the interference of confounding factors on the domain adaptation task. Our experiments conducted on four domain adaptation image classification benchmark datasets and one counterfactual dataset show that the model achi

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Performance of a four deep neural network structure in moiré fringe intelligent analysis

引用

Applied Optics 2025年第9期64卷 2208-2216页

作者： Zeng, Xinyu Chen, Yunyun Cheng, Weihao Xie, Wenzhuo Yang, Cuihong State Key Laboratory Cultivation Base of Atmospheric Optoelectronic Detection and Information Fusion Nanjing University of Information Science & Technology Nanjing210044 China Jiangsu Engineering Research Center for Intelligent Optoelectronic Sensing Technology of Atmosphere Nanjing University of Information Science & Technology Nanjing210044 China Jiangsu International Joint Laboratory on Meteorological Photonics and Optoelectronic Detection Nanjing University of Information Science & Technology Nanjing210044 China Nanjing University of Information Science & Technology Nanjing210044 China

Moiré fringe analysis holds significant importance in the assessment of flow fields, while deep learning is advancing rapidly and is frequently employed to handle complex computational challenges. In this paper, the performance of U-net, Unet++, Unet 3+, and ResUnet in moiré fringe analysis is compared and analyzed. The results indicate significant differences among the four neural networks in terms of the training time, prediction speed, prediction accuracy, and generalization capability regarding positional variations. Among the four neural networks, ResUnet exhibits the shortest training times, highest accuracy, and fastest prediction speeds while also exhibiting good generalization capability. Consequently, in practical applications, using ResUnet for moiré fringe analysis is an excellent choice. © 2025 Optica Publishing Group. All rights, including for text and data mining (TDM), Artificial Intelligence (AI) training, and similar technologies, are reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Toward High-Accuracy and Low-Latency Group Vehicle Trajectory Prediction With Linear UNet-Enhanced Fully Connected Spatial–Temporal Graph Neural Network

引用

IEEE Internet of Things Journal 2025年第9期12卷 11797-11811页

作者： Zhang, Zirui Huang, Xumin Yu, Rong Wu, Maoqiang Kang, Jiawen Xie, Shengli Guangdong University of Technology School of Automation Key Laboratory of Intelligent Information Processing and System Integration of IoT Ministry of Education Guangdong Provincial Key Laboratory of Intelligent Systems and Optimization Integration Guangzhou510006 China South China Normal University School of Electronic Science and Engineering Foshan528225 China Guangdong University of Technology School of Automation 111 Center for Intelligent Batch Manufacturing Based on IoT Technology Guangdong-Hong Kong-Macao Joint Laboratory for Smart Discrete Manufacturing Guangzhou510006 China

Group vehicle trajectory prediction (GVTP) is important for analyzing the traffic states and optimizing the traffic management. However, existing studies have performance bottlenecks in the prediction accuracy and inference latency. To tackle the problems, we propose a linear UNet-enhanced fully connected spatial–temporal GNN (LUFC-STGNN) for GVTP. First, a spatial graph is constructed by integrating prior-based and data-driven methods to capture both explicit and implicit spatial interactions between the vehicles. After that, a comprehensive temporal graph is created to capture the varying strengths of temporal interactions between all vehicles throughout historical timestamps. Furthermore, a fully connected spatial–temporal graph combining the spatial and temporal graphs is introduced to extract the effective spatial–temporal interaction features of the vehicles through the graph convolution operation. Finally, a linear UNet-based temporal dependency encoder (LU-TDE) is designed to further enhance the model’s ability of capturing the potential temporal patterns in the vehicle interactions. The encoder with linear complexity explores the multiscale temporal dependencies from the spatial–temporal interaction features but also reducing the inference latency. Experiments results based on real-world datasets show that compared to state-of-the-art models, our model reduces the average root mean square error over the 5-s prediction horizon by 31% and 10% on the NGSIM and HighD datasets, while reducing the inference latency by at least 1.26 times. © 2014 IEEE.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Infrared Dim Small Target Detection Algorithm Based on Adaptive Attention Transformer Network

引用

Journal of Shanghai Jiaotong University (Science) 2025年 1-13页

作者： Ding, Leqi Xia, Yilin Wang, Biyun Cai, Yunze Department of Automation Shanghai Jiao Tong UniversityKey Laboratory of System Control and Information Processing of Ministry of Education Shanghai200240 China National Key Laboratory of Air-based Information Perception and Fusion Henan Luoyang471009 China Shanghai Institute of Aerospace Control Technology Infrared Detection Technology R & D Center of China Aerospace Science and Technology Corporation Shanghai201109 China

This paper presents a novel algorithm for detecting infrared dim small targets in complex backgrounds by integrating both global and local feature information. The proposed method employs an adaptive attention transformer network which addresses challenges associated with computational complexity and information redundancy in traditional self-attention mechanisms. A global context aggregation module is designed to address the loss of small target features in high-level feature maps. This module compensates for spatial distribution information during the multi-scale feature fusion process, ensuring the preservation of target features. Comparison experiments were conducted on two datasets, NUAA-SIRST and IRSTD-1k. The proposed algorithm achieved 0.784 for mIoU and 0.878 for F-measure on NUAA-SIRST, 0.632 for mIoU and 0.807 for F-measure on IRST-1k, all of which are the highest values. The proposed algorithm shows a significant improvement over other existing detection algorithms, demonstrating its capability to accurately extract both local and global features of targets with high recall rates and low false alarm rates across various scenarios. The ablation study results highlight the significant improvements contributed by the key modules to the proposed algorithm. (Figure presented.) © Shanghai Jiao Tong University 2025.

关键词： A adaptive attention global context aggregation module infrared dim small target detection TP391.4 Transformer network

来源：评论

学校读者我要写书评

暂无评论

Circular Polarization Selective Metasurface for Folded Reflectarray Antenna at Ka-band

引用

IEEE Antennas and Wireless Propagation Letters 2025年第5期24卷 1278-1282页

作者： Huang, Heng Ye, Jingxin Xue, Chunhua Li, Teng Guangxi University of Science and Technology Guangxi Key Laboratory of Multidimensional Information Fusion for Intelligent Vehicles School of Electronic Engineering Liuzhou545006 China Southeast University State Key Laboratory of Millimeter Waves School of Information Science and Engineering Nanjing210096 China Purple Mountain Laboratories Nanjing211111 China

A circular polarization (CP) folded reflectarray antenna is proposed at the 30 GHz band. A five-layered circular polarization selective metasurface (CPSM) is introduced to realize the path folding for a CP source. The CPSM element consists of two back-to-back connected CP antennas with a slant I-shaped slot on the ground plane for electromagnetic wave coupling. A 2×2 patch array with a cross-shaped strip embedded in the center is designed for CP response. To enhance angular response stability, square patches are stacked on both the top and bottom sides. A metallic post through all layers is placed in the center to balance the transmission and axial ratio (AR) bandwidth. The proposed CPSM element exhibits superior AR and amplitude performance under larger-angle oblique incidence which promises the folded reflectarray antenna achieves excellent aperture efficiency along with substantial AR and gain bandwidth. To verify the proposed CPSM, a 40×40 folded reflectarray is presented. The measured results show a 3 dB AR bandwidth of 20.3% and a 3 dB gain bandwidth of 15.13%. The peak gain of 33.89 dBic is found at 30 GHz with an aperture efficiency of 60.9%. The remarkable results demonstrate the proposed CPSM for antenna engineering applications. © 2002-2011 IEEE.

关键词： Reflection

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：