检索结果-内蒙古大学图书馆

4th International Conference on Pattern Recognition and Artificial Intelligence, ICPRAI 2024

作者： Zhang, Chunhu Ibrayim, Mayire Hamdulla, Askar Deng, Qilin Xinjiang Key Laboratory of Signal Detection and Processing Urumqi China Xinjiang University School of Information Science and Engineering Urumqi China Xinjiang University College of Future Technology Urumqi China

ISBN: (纸本)9789819787012

The field of end-to-end text spotting has garnered significant interest in recent years, propelled by the revealed intrinsic synergies between scene text detection and recognition. While advancements have been made, the challenge of arbitrarily shaped scene text spotting persists. This paper introduces an innovative feature augmentation module that addresses the issues of limited receptive fields and weak representation typical of lightweight backbone networks, while also enhancing multi-scale information more effectively and reducing information loss during feature aggregation. Furthermore, to extract a richer set of backbone features, we propose a dual information attention mechanism that enables the backbone network to neuronally focus on salient information. © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Dual-Path Spectrogram Refinement Network for Robust Speaker Verification

引用

Journal of Shanghai Jiaotong University (science) 2025年 1-9页

作者： Wang, Zonghui Fang, Zhihua He, Liang School of Computer Science and Technology Xinjiang University Urumqi830017 China School of Intelligence Science and Technology Xinjiang University Urumqi830017 China Xinjiang Key Laboratory of Signal Detection and Processing Urumqi830017 China Department of Electronic Engineering Beijing National Research Center for Information Science and Technology Tsinghua University Beijing100084 China

The accuracy and reliability of automatic speaker verification (ASV) face significant challenges in noisy environments. In recent years, joint training of speech enhancement front-end and ASV back-end has been widely applied to improve the robustness of ASV systems. Traditional joint training directly uses the enhanced speech features as the input of the back-end. However, the diversity of noise types and noise intensities will excessively suppress the enhanced features, resulting in speech distortion or residual noise. To alleviate this problem, we propose a dual-path spectrogram refinement network that enables the enhanced features to learn supplementary information from the separated noise features. In addition, we incorporate coordinate attention into the overall joint architecture to capture more comprehensive frequency and temporal information of the speaker from different spatial positions. We conduct extensive experiments on the VoxCeleb1 test set, the out-of-domain noise test set, and the VOiCES corpus. The experimental results demonstrate that our proposed method significantly improves the accuracy and robustness of speaker verification systems in both clean and noisy environments. (Figure presented.) © Shanghai Jiao Tong University 2025.

关键词： Spectrographs

来源：评论

学校读者我要写书评

暂无评论

Improving Speaker Verification Back-End with Graph Neural Networks

引用

Journal of Shanghai Jiaotong University (science) 2025年 1-9页

作者： Chen, Jinfeng Fang, Zhihua He, Liang School of Computer Science and Technology Xinjiang University Urumqi830017 China Xinjiang Key Laboratory of Signal Detection and Processing Urumqi830017 China School of Intelligence Science and Technology Xinjiang University Urumqi830017 China Department of Electronic Engineering Beijing National Research Center for Information Science and Technology Tsinghua University Beijing100084 China

Currently, research on speaker verification tasks is primarily concentrated on enhancing deep speaker models to extract high-quality speaker embeddings. Nevertheless, this speaker embeddings can be regarded as potential graph structures. Therefore, this paper introduces a graph neural network-based speaker verification back-end approach, treating the speaker embeddings derived from the front-end as graph structures and leveraging graph neural networks to uncover the relationships among embeddings, thereby achieving superior-quality speaker embeddings. In addition, we propose a group updating method to solve the problem of memory overflow when the number of nodes is excessive. Extensive experiments and ablation studies were carried out on the VoxCeleb dataset, and the experimental outcomes validate the effectiveness of our proposed graph neural network speaker back-end in significantly boosting the performance of speaker verification systems. (Figure presented.) © Shanghai Jiao Tong University 2025.

关键词： Graph embeddings

来源：评论

学校读者我要写书评

暂无评论

Anomalous Sound detection Using Time-Frequency Feature and Mixbatch

引用

Journal of Shanghai Jiaotong University (science) 2025年 1-8页

作者： Huang, Shun Zhang, Yunxiang Fang, Zhihua Tang, Minrui Xu, Ruifeng He, Liang School of Computer Science and Technology Xinjiang University Urumqi830017 China Xinjiang Key Laboratory of Signal Detection and Processing Urumqi830017 China School of Intelligence Science and Technology Xinjiang University Urumqi830017 China Department of Electronic Engineering Beijing National Research Center for Information Science and Technology Tsinghua University Beijing100084 China

The sound emitted by machines under abnormal working conditions exhibits various frequency patterns. Currently, the most advanced anomalous sound detection (ASD) approach is to apply a multi-head self-attention mechanism to the Log-Mel spectrogram for automatic frequency pattern analysis. However, the Log-Mel spectrogram may filter out high-frequency components of abnormal sounds;thus the use of self-attention mechanisms on Log-Mel spectrogram seems to have certain limitations. In this paper, we construct a simple convolutional neural network to extract comprehensive frequency features from raw audio to complement spectral-temporal information fusion. The parameters of this neural network are continuously updated during the training process to extract better frequency features for downstream classification neural networks in ASD. Additionally, a method for data augmentation in the batch dimension was developed to help the classification model learn both types of comprehensive features simultaneously. The proposed method achieved an AUC of 94.13% and a pAUC of 89.09% on the DCASE 2020 Challenge Task 2 dataset. Even when using only the features proposed in this paper, an AUC of 84.62% was achieved. (Figure presented.). © Shanghai Jiao Tong University 2025.

关键词： Spectrographs

来源：评论

学校读者我要写书评

暂无评论

Infrared Small Target detection with Feature Refinement and Context Enhancement 31st

Infrared Small Target Detection with Feature Refinement and...

引用

31st International Conference on Multimedia Modeling, MMM 2025

作者： Li, Xiuhong Zhu, Xinyue Li, Boyuan Li, Songlin Wang, Luyao Jia, Zhenhong School of Computer Science and Technology Xinjiang University Urumgi China Key Laboratory of Signal Detection and Processing Xinjiang University Urumqi China College of Mathematics and System Science Xinjiang University Urumqi China

ISBN: (纸本)9789819620609

Infrared small target detection has received widespread application and attention in both civilian and military fields. However, due to the very small size and lack of unique features of these targets, existing methods often suffer from inaccurate edge localization, and the targets are easily overwhelmed by complex backgrounds. To effectively address these issues, we designed a Feature Refinement and Context Enhancement Network (FCNet). The network consists of a multi-branch feature extraction module (MFEM), which integrates ordinary convolution and center difference convolution for diversified feature extraction, and further refines features through attention mechanism to enhance feature expression ability. In addition, in order to better capture the contextual information of the target, we have introduced the Context Enhancement Module (CEM). The CEM improves the robustness and detection accuracy of the network in complex backgrounds by preserving and enhancing important information in the image. Especially when processing infrared images with complex backgrounds and low signal-to-noise ratios, CEM can effectively highlight the target area and reduce false positives and false negatives. The experimental results on the SIRST dataset show that our FCNet performs well on multiple evaluation metrics. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Infrared imaging

来源：评论

学校读者我要写书评

暂无评论

DDCTrack： Dynamic Token Sampling for Efficient UAV Transformer Tracking 27th

DDCTrack： Dynamic Token Sampling for Efficient UAV Transfor...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Du, Guocai Zhou, Peiyong Yadikar, Nurbiya Aysa, Alimjan Ubul, Kurban College of Information Science and Engineering Xinjiang University Urumqi830046 China The Key Laboratory of Multilingual Information Technology Xinjiang University Urumqi830046 China

ISBN: (纸本)9783031783531

Although state-of-the-art transformer models have shown promising results in unmanned aerial vehicle (UAV) tracking, they come with high computational demands. Existing tracking methods aim to reduce computational complexity by controlling the number of tokens. However, this method is not effective for all tracking methods. Therefore, we propose a novel dynamic token sampling for an efficient UAV transformer tracking framework. Unlike previous transformer-based tracking methods, our method avoids the need for complex head networks like classification and regression. It solely employs our newly designed encoder, comprising three key components: Dynamic Position Embedding, Dynamic Token Sampler, and Convolutional Feed-Forward Network. This module enhances visual representation by scoring and dynamically sampling tokens, allowing for a flexible token count that adapts to target changes within each frame. We utilize a simple image-sequence contrastive loss as the loss function. Our approach not only simplifies the tracking framework, but also achieves state-of-the-art performance on multiple challenging datasets at real-time run speeds. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Aircraft detection

来源：评论

学校读者我要写书评

暂无评论

RGR-KBQA: Generating Logical Forms for Question Answering Using Knowledge-Graph-Enhanced Large Language Model 31

RGR-KBQA: Generating Logical Forms for Question Answering Us...

引用

31st International Conference on Computational Linguistics, COLING 2025

作者： Feng, Tengfei He, Liang School of Computer Science and Technology Xinjiang University Urumqi830017 China Xinjiang Key Laboratory of Signal Detection and Processing Urumqi830017 China Department of Electronic Engineering Beijing National Research Center for Information Science and Technology Tsinghua University Beijing100084 China

ISBN: (纸本)9798891761964

In the field of natural language processing, Knowledge Base Question Answering (KBQA) is a challenging task that involves accurately retrieving answers from structured knowledge. Existing methods often face issues when generating query statements using LLMs, as the knowledge introduced may be imprecise and the models themselves may exhibit hallucination problems, leading to low accuracy, particularly when dealing with complex questions. To address these challenges, we introduce a novel semantic parsing approach called RGR-KBQA, which adopts a Retrieve-Generate-Retrieve framework. The first retrieval step introduces factual knowledge from a knowledge graph to enhance the semantic understanding capabilities of LLMs, thereby improving generation accuracy of logical form. The second step uses a fine-tuned model to generate the logical form, and the final step involves unsupervised relation and entity retrieval to further enhance generation accuracy. These two retrieval steps help alleviate the hallucination problems inherent in LLMs. Experimental results show that RGR-KBQA demonstrate promising performance on CWQ and WebQSP datasets. © 2025 Association for Computational Linguistics.

关键词： Question answering

来源：评论

学校读者我要写书评

暂无评论

Stable Extended U-Net for Noise-Robust Speaker Verification

Stable Extended U-Net for Noise-Robust Speaker Verification

引用

2025 IEEE International Conference on Acoustics, Speech, and signal Processing, ICASSP 2025

作者： Wang, Zonghui Fang, Zhihua He, Liang School of Computer Science and Technology Xinjiang University Urumqi830017 China School of Intelligence Science and Technology Xinjiang University Urumqi830017 China Xinjiang Key Laboratory of Signal Detection and Processing Urumqi830017 China Department of Electronic Engineering Tsinghua University Beijing100084 China Beijing National Research Center for Information Science and Technology Tsinghua University Beijing100084 China

ISBN: (纸本)9798350368741

With advancements in deep learning, speaker verification systems have significantly improved their performance in noisy environments. Researchers typically demonstrate the effectiveness of their improved models by comparing performance on specific datasets, such as the VoxCeleb benchmark. However, in diverse real-world noise conditions, the out-of-domain generalization ability is also a crucial factor in evaluating a model's performance improvement. Research on stable learning indicates that eliminating the spurious correlation between training and testing data can enhance the generalization of the model. Building on this idea, we propose an improved speaker verification system with high generalization based on the extended U-Net (ExU-Net). It uses the sample reweighting method from stable learning to eliminate sample correlations and retains more effective speaker information through sub-pixel convolutions and coordinate attention mechanisms. We validate the effectiveness of this approach through extensive evaluations on VoxCeleb1, VOiCES, and other out-of-domain noise test sets, highlighting its generalization capability and model robustness. © 2025 IEEE.

关键词： noisy environments out-of-domain generalization Speaker verification stable learning

来源：评论

学校读者我要写书评

暂无评论

MCFTNet: Multimodal Cross-layer Fusion Transformer Network for Hyperspectral and LiDAR Data Classification

引用

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 2025年 18卷 12803-12818页

作者： Huang, Wei Wu, Tianren Zhang, Xueyu Li, Liangliang Lv, Ming Jia, Zhenhong Zhao, Xiaobin Ma, Hongbing Vivone, Gemine Xinjiang University School of Computer Science and Technology Urumqi830017 China Xinjiang University Key Laboratory of Signal Detection and Processing Urumqi830017 China Guangxi University School of Computer and Electronic Information Nanning530004 China Guangxi University Guangxi Key Laboratory of Multimedia Communications and Network Technology Nanning530004 China Beijing Institute of Technology School of Information and Electronics Beijing100081 China University of Science and Technology Beijing School of Computer and Communication Engineering Beijing100083 China Tsinghua University Department of Electronic Engineering Beijing100084 China National Research Council Tito85050 Italy NBFC National Biodiversity Future Center Palermo90133 Italy

Remote sensing image classification is a popular yet challenging field. Many researchers have combined convolutional neural networks (CNNs) and Transformers for hyperspectral image (HSI) classification tasks. However, in traditional Transformers, shallow-level information does not propagate well to deeper layers, which can lead to spatial variations and overfitting. Moreover, traditional Transformer models use an external classification (CLS) token that is randomly initialized and often struggles to generalize effectively. In this paper, we combine the strengths of HSI and light detection and ranging (LiDAR) data, using LiDAR as an external CLS token, which significantly enhances classification accuracy and reliability. We propose a new multimodal cross-layer fusion transformer network (MCFTNet), integrating CNNs with the latest Transformer networks. It includes a CNN for extracting spatial features and a hybrid cross-patch attention mechanism (HyCrossAttention) for land cover classification, leveraging LiDAR data to generate CLS and HSI patch tokens. More importantly, to reduce the loss of valuable information during the layer-by-layer propagation, we designed cross-layer skip connections. Through adaptive learning, these cross-layer fusions help address the gradient vanishing problem in deep networks while preserving early-layer features. This enables the model to better integrate information from different layers, enhancing both its stability and performance. We carried out in - depth experiments on commonly - used benchmark datasets, specifically, the University of Houston dataset, the Trento dataset, and the University of Southern Mississippi Gulf Park (MUUFL) dataset. We compared the results of the proposed MCFTNet model with those obtained from state-of-the-art Transformer models, classical CNNs, and traditional classifiers. As a result, the MCFTNet model outshone them all in terms of performance. © 2008-2012 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

High precision and low noise demodulation scheme using adaptive fast fourier transform for fiber-optic interferometric sensors

引用

Optics Express 2025年第1期33卷 817-828页

作者： Liu, Zheng Lou, Ruizhe Zhang, Lingling Feng, Lei Wang, Fuyin Yao, Qiong Wu, Weijun Xia, Ji College of Meteorology and Oceanography National University of Defense Technology Changsha410073 China Hunan Key Laboratory for Marine Detection Technology Changsha410073 China College of Information Science and Engineering Northeastern University Shenyang110819 China Hunan Institute of Advanced Technology Changsha410072 China

This paper presents an adaptive fast Fourier transform (adaptive FFT) demodulation scheme, aimed at enhancing the precision and noise suppression capability of signal processing in fiber-optic interferometric sensors. By adaptively optimizing the length of the acquired spectrum and dynamically adjusting the frequency domain resolution, the proposed scheme can precisely calculate the eigenfrequency of the reflected spectrum. Therefore, the adaptive FFT demodulation scheme can effectively enhance the extraction ability of phase quadrature demodulation signal. In comparison to traditional FFT demodulation scheme, the proposed scheme demonstrates nearly a tenfold improvement in eigenfrequency calculation accuracy. The experimental results indicate a significant improvement in the stability of the adaptive FFT demodulation scheme. The fluctuations are reduced to approximately 0.03 dB. This value is roughly equivalent to 25% of the fluctuations typically observed in traditional FFT demodulation schemes. Additionally, the adaptive FFT scheme demonstrates superior adaptability to sensors with varying optical path differences (OPDs). It maintains noise levels below -100 dB@1 kHz across sensors with different OPDs, and substantially reduces noise fluctuations compared to traditional FFT scheme. Furthermore, the proposed scheme exhibits notable advantages in crosstalk suppression for multiplexed signal demodulation. The fundamental frequency crosstalk (FFC) is reduced to below -50 dB, which is approximately 10 dB lower than that achieved by traditional FFT methods. The adaptive FFT demodulation scheme shows promising potential for applications in low noise, high precision, and dynamic signal detection within fiber-optic interferometer sensors and multiplexing arrays. © 2025 Optica Publishing Group (formerly OSA). All rights reserved.

关键词： Frequency domain analysis

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：