检索结果-内蒙古大学图书馆

Multi-Level Interactive Fusion Network Based on Adversarial Learning for Fusion Classification of Hyperspectral and Lidar Data

学校读者我要写书评

暂无评论

SSRN 2024年

作者： Fan, Yingying Qian, Yurong Gong, Weijun Chu, Zhuang Qin, Yugang Muhetaer, Palidan College of Information Science and Engineering Xinjiang University Xinjiang Urumqi China College of Software Xinjiang University Xinjiang Urumqi China Key Laboratory of Signal Detection and Processing in Xinjiang Uygur Autonomous Region Xinjiang Urumqi China Key Laboratory of Software Engineering Xinjiang Urumqi China College of Information Management Xinjiang University of Finance and Economics Xinjiang Urumqi China

With the growing awareness of the limitations of unimodal data, research on joint classification usingmultimodal remote sensing data has garnered increasing attention. However, existing methods stillhave certain deficiencies in dealing with feature extraction and feature fusion of multimodal remotesensing data. In order to obtain more accurate classification results, it is crucial to fully exploit andutilize the complementary information in multimodal remote sensing data. In this paper, a multi-levelinteractive fusion network based on adversarial learning (MLIF-AL) is proposed for Hyperspectralimage (HSI) and light detection and ranging (LiDAR) data fusion classification. First, the featureextraction part based on generative adversarial network (GAN) utilizes adversarial training betweenthe generator and the discriminator to guide the network to learn more discriminative and distinguishedfeature representations. meanwhile, a cross-modal interactive information extraction (CMIIE) moduleis also designed for the generator encoding part to promote the interaction between multimodal datafrom a global perspective and realize the full utilization of multimodal complementary information. Inaddition, in the multi-level feature fusion classification (MLFFC) part, the complementarities betweenthe features learned by the neural networks at each layer are comprehensively utilized to obtain higherlevel multimodal fusion semantic information. Finally, the adversarial loss and classification loss arecombined for network training. Extensive experiments on three commonly used HSI and LiDARdatasets and comparisons with state-of-the-art methods demonstrate the effectiveness and superiorityof the model. © 2024, The Authors. All rights reserved.

关键词： Remote sensing

Sle Diagnosis Research Based on Sers Combined with a Multi-Modal Fusion Method

学校读者我要写书评

暂无评论

SSRN 2023年

作者： Huang, Yuhao Chen, Chen Chang, Chenjie Cheng, Zhiyuan Liu, Yang Chen, Cheng Lv, Xiaoyi College of Software Xinjiang University Xinjiang 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 Xinjiang Cloud Computing Engineering Technology Research Center Karamay834000 China

As artificial intelligence technology gains widespread adoption in biomedicine, the exploration of integrating biofluidic Raman spectroscopy for enhanced disease diagnosis opens up new prospects for the practical application of Raman spectroscopy in clinical settings. However, for Systemic Lupus Erythematosus (SLE), the relatively weak signals from a single Raman spectroscopic data make it difficult to obtain desirable classification results. Although the surface enhancement technique can enhance the scattered signal of Raman spectroscopic data, the sensitivity of the SERS substrate to airborne impurities and the non-uniform distribution of hotspots degrade some of the signals. To fully utilize both types of data, this paper proposes a two-branch residual-attention network (DBRAN) fusion technique that allows the original spectra to complement the degraded portion, improving the model's classification accuracy. The residual module is used to extract features, which preserves the original features while extracting the deep ones. At the same time, the study incorporates the attention module in both the upper and lower branches to handle the weight assignment of the two modal features more efficiently. The experimental results demonstrated that both the low-level fusion method and the intermediate-level fusion method significantly improved the diagnostic accuracy of SLE disease classification when compared with a single modality, and the intermediate-level fusion DBRAN achieved 100% classification accuracy, sensitivity, and specificity. Compared with the raw Raman spectrum single mode and SERS single mode, the improvement is 10% and 7%, respectively. This experiment realized the rapid diagnosis of SLE disease by fusing the spectral data of multiple modes, which provides a reference idea in the Raman spectroscopy field. © 2023, The Authors. All rights reserved.

关键词： Raman spectroscopy

Sle Diagnosis Research Based on Sers Combined with a Multi-Modal Fusion Method

学校读者我要写书评

暂无评论

SSRN 2023年

关键词： Raman spectroscopy

Asffuse: Infrared and Visible Image Fusion Model Based on Adaptive Selection Feature Maps

学校读者我要写书评

暂无评论

SSRN 2023年

作者： Liu, Kuizhuang Li, Min Zuo, Enguang Chen, Chen Chen, Cheng Wang, Bo Lv, Yi Xiao Wang, Yunling College of Software Xinjiang University XinJiang Urumqi830046 China College of Information Science and Engineering Xinjiang University XinJiang Urumqi830046 China Key Laboratory of Signal Detection and Processing Xinjiang University XinJiang Urumqi830046 China Department of Radiology The First Affiliated Hospital of Xinjiang Medical University XinJiang Urumqi830000 China

Scientists are constantly modifying the architecture of deep learning networks to achieve better fusion results, but few people pay attention to the impact of noise feature maps generated in the convolution process on fusion results. Here, we aim to minimize the influence of noisy feature maps on fusion results and propose a fusion model, namely infrared and visible image fusion model based on adaptive selection feature maps (ASFFuse), which incorporates an adaptive selection feature maps module (ASFM). The ASFM effectively filters out noisy feature maps to enhance the fusion performance. Additionally, we introduce a feature enhancement module (FEM) to enrich fusion image with more source image information. For unsupervised training of the proposed model, We propose a gradient loss function based on contrast learning. This loss function can keep the texture information we need and discard the texture information we do not want. By using this gradient loss function, our fused images can have more gradient information and discard unwanted texture information, which makes the fusion images have better visual effects. Our ASFFuse model has been shown to outperform state-of-the-art models in both quantitative and qualitative evaluations in extensive experiments on TNO and RoadScene datasets. Code is available at https://***/LKZ1584905069/ASFFuse. © 2023, The Authors. All rights reserved.

关键词： Image fusion

A Fiber Vibration signal Identification Method Based on the Combination of EWT-FE and LSTM

学校读者我要写书评

暂无评论

Journal of Physics: Conference Series 2021年第1期2112卷

作者： Xin Zhang Qingmo Ja SaiSai Ruan Qin Hu Key Laboratory of Signal Detection and Processing College of Information Science and Engineering Xinjiang University Ürümqi 830046 China

As the optical fiber perimeter security system is widely used in real life, how to identify the types of intrusion events in a timely and effective manner is becoming a major research hotspot. At present, in this field, various signal feature extraction algorithms are usually used to extract intrusion signal features to form feature vectors, and then machine learning algorithms are used to classify the feature vectors to achieve the role of identifying the types of intrusion events. As a common signal feature extraction algorithm, the EMD algorithm has been widely used in the feature extraction of various vibration signals, but it will have the problem of modal aliasing and affect the feature extraction effect of the signal. Therefore, EWT, VMD and other algorithms have been successively used proposed to improve modal aliasing. On the basis of fully comparing the existing algorithms, this paper proposes a fiber vibration signal identification method that decomposes the signal through the empirical wavelet transform (EWT) algorithm and then extracts the fuzzy entropy (FE) of each component, and uses LSTM for classification. The final experiment shows that the method can identify four kinds of fiber intrusion signals in time and effectively, with an average recognition accuracy rate of 97.87%, especially for flap and knock recognition rate of 100%.

关键词：

Research on Coordinated Coverage of Non-Lambertian Optical Wireless Communications

学校读者我要写书评

暂无评论

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

International Conference on Computer and Communications (ICCC)

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

ISBN: (纸本)9781665409513

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.

关键词： Wireless communication Integrated optics Optical losses Numerical analysis Optical switches Optical fiber communication Optical transmitters

Secure Optical Wireless Links with Dynamic Beam and Diversity Receiver

学校读者我要写书评

暂无评论

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

IEEE International Conference on Network Infrastructure and Digital Content (IC-NIDC)

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

ISBN: (纸本)9781665405836

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.

关键词： Wireless communication Integrated optics Measurement Conferences Optical fiber networks Physical layer security Optical receivers