检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Hu, Ying Zhu, Xiujuan Li, Yunlong Huang, Hao He, Liang Key Laboratory of signal detection and processing in Xinjiang China School of Information Science and Engineering Xinjiang University Urumqi China Department of Electronic Engineering Tsinghua University China

Sound event detection (SED) is an interesting but challenging task due to the scarcity of data and diverse sound events in real life. This paper presents a multi-grained based attention network (MGA-Net) for semi-supervised sound event detection. To obtain the feature representations related to sound events, a residual hybrid convolution (RH-Conv) block is designed to boost the vanilla convolution's ability to extract the time-frequency features. Moreover, a multi-grained attention (MGA) module is designed to learn temporal resolution features from coarse-level to fine-level. With the MGA module, the network could capture the characteristics of target events with short- or long-duration, resulting in more accurately determining the onset and offset of sound events. Furthermore, to effectively boost the performance of the Mean Teacher (MT) method, a spatial shift (SS) module as a data perturbation mechanism is introduced to increase the diversity of data. Experimental results show that the MGA-Net outperforms the published state-of-the-art competitors, achieving 53.27% and 56.96 % event-based macro F1 (EB-F1) score, 0.709 and 0.739 polyphonic sound detection score (PSDS) on the validation and public set respectively. Copyright © 2022, The Authors. All rights reserved.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Detecting Video Image Changes Based on Improved Difference Map and Image Fusion

Detecting Video Image Changes Based on Improved Difference M...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Ang Tian Zhenhong Jia Xiaohui Huang Sensen Song Jiajia Wang Gang Zhou Fei Shi School of Computer Science and Technology Xinjiang University Xinjiang China Key Laboratory of Signal Detection and ProcessingXinjiang University Xinjiang China

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

To resolve the problems of complex random noise and poor image quality in hawk-eye surveillance video images under low illumination, a surveillance video image change detection method based on K-means clustering and differential image adaptive fusion was developed. First, the log ratio operator and extreme average ratio operator were used to obtain two difference maps for two video surveillance images at different times. Then, the Laplacian pyramid was used for adaptive fusion. Next, the TVL1 dual method was used to denoise the fused image, PCA was used to fuse the denoised image and the enhanced image of the changed region, and then the adaptive median filter was used for filtering. Finally, the two-stage clustering method was used to obtain the final change detection results for the difference map. Based on our experimental findings, our method outperforms comparative algorithms in many metrics, including accuracy, robustness, time consumption, and false alarm rate.

关键词： Laplace equations Fuses Noise Lighting Clustering algorithms Adaptive filters Filtering algorithms

来源：评论

学校读者我要写书评

暂无评论

Secure Optical Wireless Links with Dynamic Beam and Diversity Receiver 7

Secure Optical Wireless Links with Dynamic Beam and Diversit...

引用

7th IEEE International Conference on Network Intelligence and Digital Content, IC-NIDC 2021

作者： DIng, Jupeng Lin, Chih I Wang, Jintao Yang, Hui Wang, Lili School of Information Science and Engineering Xinjiang University Key Laboratory of Signal Detection and Processing in Xinjiang Uygur Autonomous Region China China Mobile Research Institute Beijing China Tsinghua University Beijing National Research Center for Information Science and Technology Department of Electronic Engineering Beijing China Ludong University School of Information and Electrical Engineering Yantai China

ISBN: (纸本)9781665405829

Optical wireless communication (OWC) has emerged as one promising candidate to mitigate the frequency spectrum crisis. Due to the broadcast nature of OWC channel, physical layer security (PLS) techniques have been considered and explored to improve the transmission confidentiality of OWC links. However, almost all current schemes focus on scenarios including multiple transmitters and fail to match the scenarios with limited transmitters, even single transmitter. For addressing this issue, in this work, the dynamic optical beam based PLS enhancement scheme is proposed. Unlike conventional Lambertian beam based technique paradigm, the above scheme tentatively utilize the commercially available non-Lambertian beam to configure the secure OWC links. Numerical results show that, compared with about 1.10 bps/Hz average secrecy capacity (SC) of the conventional configuration, up to 1.93 bps/Hz average SC gain could be provided by the proposed dynamic beam scheme. Moreover, this potential gain will be increased to about 2.39 bps/Hz when the diversity combing is available for the legitimate user receiver. © 2021 IEEE.

关键词： Dynamics

来源：评论

学校读者我要写书评

暂无评论

MP3Net:Multi-scale Patch Parallel Prediction Networks for Multivariate Time Series Forecasting

MP3Net:Multi-scale Patch Parallel Prediction Networks for Mu...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Daishun Cui Jiwei Qin Fang He Fei Shi Qiang Li Dezhi Sun Jiachen Xie School of Computer Science and Technology Xinjiang University Urumqi China Xinjiang Key Laboratory of Signal Detection and Processing Xinjiang University Urumqi China HongYou Software Co. Ltd Karamay China

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

Transformers have emerged as a crucial technology in long-term time series forecasting (LTSF). Recent studies have demonstrated that Transformers segment time series into patches, effectively capturing temporal patterns. However, single-scale patches still fail to adequately capture the complex short-term fluctuations and long-term trends in time series. To tackle this issue, we propose a novel framework, the Multi-scale Patch Parallel Prediction Network (MP3Net). Firstly, we propose a multi-scale patch module to extract local features and long-term correlations from the time series to effectively capture and identify data characteristics across different time scales. Secondly, we propose an adaptive fusion module to dynamically balance the weight of local detail features and long- term correlations from each branch. Thirdly, we introduce a hybrid loss function for model training aimed at enhancing the sensitivity and adaptability of the network towards different types of errors, consequently improving overall performance. Experiments conducted on five benchmark datasets demonstrate that MP3Net surpasses existing methods in LTSF and MP3Net presents an effective approach for addressing the challenge of modeling long-term dependencies in time series analysis.

关键词： Training Adaptation models Correlation Sensitivity Time series analysis Neural networks Transformers

来源：评论

学校读者我要写书评

暂无评论

Rapid diagnosis of osteoarthritis using serum Raman spectroscopy based on MFCC transformation combined with MCNN 4

Rapid diagnosis of osteoarthritis using serum Raman spectros...

引用

4th International Conference on Optics and Image processing, ICOIP 2024

作者： Xu, Hongze Chen, Chen Tian, Xuecong He, Liang Zuo, Enguang Chen, Xiaomei Wu, Lijun Chen, Cheng Lv, Xiaoyi College of Software Xinjiang University Urumqi830046 China College of Information Science and Engineering Xinjiang University Urumqi830046 China Key Laboratory of signal detection and processing Xinjiang University Urumqi830046 China Xinjiang Cloud Computing Application Laboratory Karamay834099 China Department of Rheumatology and Immunology People's Hospital of Xinjiang Uygur Autonomous Region Urumqi830001 China Xinjiang Clinical Research Center for Rheumatoid Arthritis Urumqi830001 China Department of Electronic Engineering Tsinghua National Laboratory for Information Science and Technology Tsinghua University Beijing China School of Information Science and Engineering Xinjiang University Ürümqi China

ISBN: (纸本)9781510682481

Osteoarthritis (OA) is a common chronic degenerative disease worldwide. Currently, traditional diagnostic methods for OA have disadvantages such as high cost and long time consumption. For this reason, it is urgent to have a method that can diagnose osteoarthritis quickly and accurately. During this study, we suggest a multi-column convolutional neural networks (MCNN) model based on Mel-frequency cepstral coefficient (MFCC). We first compare the effectiveness of Partial Least Squares (PLS) and MFCC in preprocessing the data and comprehensively evaluate the performance of the model by five-fold cross-validation. To further improve the ability of recognizing spectral features, We utilize the Spectral Segmentation Method (SSM) to construct the MFCC feature sequences into 2D spectral feature maps and input them into the MCNN for classification. The final experimental results show that our proposed mfc - ssm - mcnn model has a high classification efficiency in diagnosing osteoarthritis with an accuracy of 95.0%. © 2024 SPIE.

关键词： Raman spectroscopy

来源：评论

学校读者我要写书评

暂无评论

Research on Disease Diagnosis Based on Teacher-Student Network and Raman Spectroscopy

SSRN

引用

SSRN 2023年

作者： Chen, Zishuo Tian, Xuecong Chen, Chen Lv, Xiaoyi Chen, Cheng College of Software Xinjiang University Xinjiang Urumqi830046 China College of Information Science and Engineering Xinjiang University Xinjian Urumqi830046 China Key Laboratory of Signal Detection and Processing Xinjiang University Xinjiang Urumqi830046 China

In the field of spectral analysis, the common Raman spectral feature selection model can extract features effectively, but it will change the original data *** teacher model assists the student model in distillation training by transferring the trained knowledge to the student *** this process, teacher models provide guidance and help student models better understand and master complex knowledge *** teacher-student network combined with Raman spectroscopy can perform feature selection on the premise of retaining the original features, and transfer the performance of the complex deep neural network structure to another lightweight network structure *** the construction of teacher-student network, this study of diabetic nephropathy and primary Sjogren's syndrome to verify, select Isomap, t-SNE, LLE, MDS, SE five models as the teacher network, build MLP neural network as the student network, multi-layer perceptron classification, and according to the evaluation index accuracy, precision, recall, F1 index to select the best *** 50% cross-validation, the results showed that the accuracy of t-SNE and SE models in the diagnosis of diabetic nephropathy was up to 98.3%, which was 14.02% higher than that of existing *** the diagnosis of primary Sjogren's syndrome, the accuracy of other models except MDS was 100%, which was 10.48% higher than that of existing ***, Raman spectroscopy combined with teacher-student networks for disease diagnosis is an efficient and accurate protocol that has produced good experimental results in the diagnosis of diabetic nephropathy and Sjogren's *** study provides a new way of thinking for the diagnosis of other diseases. © 2023, The Authors. All rights reserved.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Research on Coordinated Coverage of Non-Lambertian Optical Wireless Communications 7

Research on Coordinated Coverage of Non-Lambertian Optical W...

引用

7th International Conference on Computer and Communications, ICCC 2021

作者： Ding, Jupeng Chih-Lin, I. Wang, Jintao Yang, Hui Wang, Lili Xinjiang University Key Laboratory of Signal Detection and Processing In Xinjiang Uygur Autonomous Region School of Information Science and Engineering Ürümqi China China Mobile Research Institute Beijing China Tsinghua University Beijing National Research Center For Information Science and Technology Department of Electronic Engineering Beijing China Ludong University School of Information and Electrical Engineering Yantai China

ISBN: (纸本)9781665409506

To a large extent, the performance of wireless optical communication system is affected by the light source radiation characteristics. Previous works are limited to rotationally symmetric Lambertian beam configurations, which fail to support target location oriented coordinated coverage. Different from the existing research, this paper proposes a coordinated coverage enhancement scheme based on asymmetric non-Lambertian optical beam configuration. Without loss of generality, this paper presents numerical analysis of coordinated and non-coordinated coverage performance in typical medium-sized indoor application scenario. Compared with non-coordinated coverage, the numerical results illustrate that the proposed coordinated coverage scheme could introduce about 4.08 dB SNR gain without transmission power increase. Therefore, non-Lambertian beam cooperation should be viewed as one efficient wireless optical communication technology. © 2021 IEEE.

关键词： Light sources

来源：评论

学校读者我要写书评

暂无评论

Research on Disease Diagnosis Based on Teacher-Student Network and Raman Spectroscopy

SSRN

引用

SSRN 2023年

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

CosDiff: Code-Switching TTS Model Based on A Multi-Task DDIM

CosDiff: Code-Switching TTS Model Based on A Multi-Task DDIM

引用

IEEE International Conference on Multimedia and Expo (ICME)

作者： Ke Chen Zhihua Huang Kexin Lu Yonghong Yan School of Computer Science and Technology Xinjiang University Urumqi China Xinjiang Key Laboratory of Signal Detection and Processing Xinjiang University Urumqi China Chinese Academy of Sciences Key Laboratory of Speech Acoustics and Content Understanding Institute of Acoustics Beijing China

ISBN: (数字)9798350390155

ISBN: (纸本)9798350390162

Although existing Text-To-Speech (TTS) synthesizers are able to generate high-quality speech in most cases, their overall performance is still affected by the distribution of the training data. When processing tasks that involve complex data distributions, such as code-switching TTS, these models might generate speech that sounds unnatural or has low speaker similarity. In this paper, we propose CosDiff, a Code-Switching TTS model based on a multi-task Denoising Diffusion Implicit Model (DDIM), which integrates Voice Conversion (VC) and TTS functionalities. Utilizing the VC function, we construct a single-speaker bilingual dataset for training, achieving a superior code-switching synthesis performance compared to the outcomes of the speaker encoder method, which trains with multiple single-speaker monolingual datasets. In addition, we employ strategies of directly predicting the clean data x 0 and progressive diffusion distillation, further accelerating the model’s sampling process. The experimental results demonstrate the efficacy of this method in improving the quality of generation, increasing sampling speed, and distilling the model.

关键词： Training Speech coding Synthesizers Training data Multitasking Probabilistic logic Data models Text to speech Speech processing Spectrogram

来源：评论

学校读者我要写书评

暂无评论

RFBiCF: A Relation-First Bidirectional Cascade Framework for Relational Triple Extraction

RFBiCF: A Relation-First Bidirectional Cascade Framework for...

引用

Pattern Recognition and Machine Learning (PRML), IEEE International Conference on

作者： Deng Jinxin Qin Xizhong Yang Rong Lv Xiaoyi School of Information Science and Engineering Xinjiang University of China Urumqi China Xinjiang Key Laboratory of Signal Detection and Processing Urumqi China School of Information The Fifth Affiliated Hospital of Xinjiang Medical Urumqi China

Relational triple extraction from unstructured text, which is in form of (subject, relation, object), is crucial in information extraction and knowledge graph construction. However, most existing methods still suffer from some limitations, such as error propagation, redundancy of relation prediction, and poor generalization. To overcome these deficiencies, we propose a relation-first bidirectional cascade multi-task framework for relational triple extraction. This framework solves the relational redundancy problem by relation prediction, alleviates error propagation through a complementary bidirectional entity tagger, and uses multi-task learning to improve the generalization of the model. Concretely, we first use a relation detection component to predict potential relations in sentences and conduct the following entity pairs extraction based on specific relations. Then, we extract entity pairs from two paralleled directions. Meanwhile, entity recognition is used as an auxiliary task, improving generalization capability. Extensive experiments were conducted on the NYT and WebNLG datasets We evaluate our model on four public datasets to demonstrate the effectiveness of our proposed RFBiCF framework, even in the complex scenario of overlapping triples.

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：