检索结果-内蒙古大学图书馆

Deep Learning Super-Resolution-Based Channel Completion for Massive MISO Systems

IEEE signal processing LETTERS 2025年 32卷 2254-2258页

作者： Zu, Keke He, Yuhan Chen, Hongyang Zheng, Yu Haardt, Martin Univ Elect Sci & Technol China Yangtze Delta Reg Inst Quzhou Quzhou 324000 Zhejiang Peoples R China Shenzhen Univ Shenzhen Key Lab Adv Machine Learning & Applicat Shenzhen 518060 Peoples R China Zhejiang Lab Res Ctr Graph Comp Hangzhou 311121 Peoples R China JD Intelligent Cities Res Beijing 100176 Peoples R China Ilmenau Univ Technol Commun Res Lab D-98693 Ilmenau Germany

With the deployment of large-scale antenna arrays, the already limited time-frequency resources are becoming increasingly scarce. In this study, we propose a novel Laplacian Pyramid Channel Completion Network (LPCCNet) designed for channel completion, thereby reducing the demand for time-frequency resources in massive MIMO systems. Compared with existing network models, the proposed LPCCNet, by employing a progressive upsampling architecture, effectively mitigates aliasing effects, suppresses error propagation, and achieves a substantial reduction in computational complexity. The simulation results show that LPCCNet achieves a superior channel completion quality compared to existing methods, particularly in rapidly time-varying scenarios.

关键词： Channel estimation Feature extraction image reconstruction Interpolation Training Superresolution Filters Vectors Convolutional neural networks Time-frequency analysis Channel completion deep learning super-resolution

来源：评论

学校读者我要写书评

暂无评论

Synergistic spectral and spatial feature analysis with transformer and convolution networks for hyperspectral image classification

引用

signal image AND VIDEO processing 2024年第4期18卷 2975-2990页

作者： Yadav, Dhirendra Prasad Kumar, Deepak Jalal, Anand Singh Kumar, Ankit Kada, B. GLA Univ Dept Comp Engn & Applicat Mathura UP India NIT Meghalaya Dept Comp Engn Shillong Meghalaya India Guru Ghasidas Vishwavidyalaya Dept Informat Technol Bilaspur Chattisgarh India King Abdulaziz Univ Dept Aerosp Engn Jeddah Saudi Arabia

Hyperspectral imaging (HSI) contains several land cover objects with rich spatial and spectral features. By utilizing these features, deep convolution neural networks (CNN) improved HSI classification accuracy. However, shallow CNN lacks global co-relation of the spatial and spectral features. Further, by increasing the convolution layers, trainable parameters also increase. Hence, computation cost significantly increases. In this study, a fusion-based HFTNet model is designed that extracts features via convolution and transformer block to improve classification performance. In the proposed HFTNet, the convolution block extracts local semantic features, and the transformer block captures the attention-based global features. We reduced the computation costs by dividing the query vector into two parts and passing it to convolution and transformer blocks for feature extraction. Finally, features are combined to generate enhanced semantic local and global features. The effectiveness of the proposed method is tested on four datasets and achieved an accuracy of 99.34% (UP), 97.95% (IP), 99.70% (SV), and 84.23% (KSC). We found that HFTNet takes less computation time and achieves much better classification accuracy than other methods.

关键词： Hyperspectral imaging (HSI) CNN Semantic features Classification Transformer

来源：评论

学校读者我要写书评

暂无评论

Three-dimensional Softmax Mechanism Guided Bidirectional GRU Networks for Hyperspectral Remote Sensing image Classification

引用

signal processing 2023年第1期212卷

作者： Wu, Guoqian Ning, Xin Hou, Luyang He, Feng Zhang, Hengmin Shankar, Achyut Beijing Inst Technol Sch Mechatron Engn Beijing Peoples R China China Aerosp Sci & Technol Corp Intelligent Unmanned Syst Overall Technol Res & De Beijing Peoples R China Chinese Acad Sci Inst Semicond High Speed Circuits & Artificial Neural Networks L Beijing 100083 Peoples R China Beijing Univ Posts & Telecommun Sch Comp Sci Natl Pilot Software Engn Sch Beijing 100876 Peoples R China Univ Sci & Technol China Hefei 230026 Peoples R China Nanyang Technol Univ NTU Sch Elect & Elect Engn Singapore Singapore Univ Warwick WMG Coventry CV7 4AL England Graph Era Deemed Be Univ Dept Comp Sci & Engn Dehra Dun 248002 India Lovely Profess Univ Sch Comp Sci Engn Phagwara 144411 Punjab India

Hyperspectral data is a valuable source of both spectral and spatial information. However, to enhance the classification accuracy of hyperspectral image features, it is crucial to capture the spatial spectral features of image elements. The recent years have witnessed the potentials of deep learning methods have shown great promise in the hyperspectral image classification due to their ability to model complex structures and extract multiple features in an end-to-end fashion. Since hyperspectral images can be viewed as sequential data, we propose a novel three-dimensional Softmax mechanism-guided bidirectional GRU network (TDS-BiGRU) for HSI classification. By utilizing a bidirectional GRU to process the sequence data, our method can significantly reduce the processing time. Furthermore, the proposed three-dimensional Softmax mechanism leverages three branches to capture cross-latitude interactions and calculate Softmax weights, which enables us to obtain deeper features with greater discriminative power. The experimental results demonstrate that the proposed method outperforms several prevalent algorithms on four hyperspectral remote sensing datasets. Additionally, we conduct thorough comparisons and ablation tests, which further confirm the effectiveness of our approach. & COPY;2023 Elsevier B.V. All rights reserved.

关键词： Three-dimensional Softmax Mechanism neural Network Hyperspectral image Classification Remote Sensing image

来源：评论

学校读者我要写书评

暂无评论

Integrating artificial intelligence and holographic imaging for advanced cervical cancer diagnosis

引用

signal image AND VIDEO processing 2025年第7期19卷 1-30页

作者： Nazir, Asifa Hussain, Ahsan Singh, Mandeep Assad, Assif Islamic Univ Sci & Technol Ctr Artificial Intelligence Pulwama 192122 J&K India Islamic Univ Sci & Technol Dept Phys Pulwama 192122 J&K India

This study investigates a novel approach to cervical cancer diagnosis by applying "Artificial Intelligence (AI)" to holographic imaging data. Specifically, it investigates AI's role in medical image analysis through "Machine Learning (ML)" and "Deep Learning (DL)", with an emphasis on holography microscopic imaging as an emerging diagnostic technique. An analysis of previous studies revealed a significant gap in exploring the potential of holographic imaging for cervical cancer diagnosis using DL approaches. This study uniquely applies Explainable AI (XAI) methods to holography microscopic imaging data for cervical cancer, filling a gap no prior research has addressed. A thorough experimentation using ML and DL approaches is done to bridge this research gap and improve diagnostic accuracy. Various pre-trained DL approaches, including "DenseNet121", "Xception", "InceptionV3", "VGG-16", "ResNet50" and "EfficientNetB4", were utilized for feature extraction. These extracted features were then classified using algorithms such as "Support Vector Machines (SVM)", "Random Forest (RF)", "K-Nearest Neighbors (KNN)","Decision Trees (DT)", "AdaBoost" and "Gradient Boosting Machine (GBM) to enhance diagnostic accuracy. The combination of DenseNet121 and SVM achieved the highest performance across all metrics, attaining 100% accuracy, precision, recall, F1-score, and AUC-score (1.00). Additionally, it recorded a Mean Squared Error (MSE) of 0.00 and a Root Mean Squared Error (RMSE) of 0.00, indicating perfect classification performance. Moreover, cross-validation on both the original dataset and the Synthetic Minority Over-sampling Technique-augmented dataset exhibited improved performance. The augmented dataset achieved higher accuracy, precision, and recall, effectively addressing class imbalance thereby enhancing classification. 5-fold cross-validation surpassed 3-fold, which demonstrated moderate accuracy (66.67-75%), high losses (0.36-0.69), and inconsistent metrics, with som

关键词： Cervical cancer Convolutional neural networks Deep learning Early diagnosis Gradient-weighted class activation mapping Holography

来源：评论

学校读者我要写书评

暂无评论

Highly applicable and imperceptible watermark attack network

引用

signal processing 2025年 230卷

作者： Wang, Chunpeng Tian, Pengfei Xia, Zhiqiu Li, Qi Li, Jian Wei, Ziqi Luo, Ting Ma, Bin Qilu Univ Technol Shandong Acad Sci Shandong Comp Sci Ctr Minist EducKey Lab Comp Power Network & Informat Jinan 250353 Peoples R China Shandong Fundamental Res Ctr Comp Sci Shandong Prov Key Lab Comp Networks Jinan Peoples R China Chinese Acad Sci Inst Automat Beijing 100190 Peoples R China Ningbo Univ Coll Sci & Technol Ningbo 315000 Peoples R China

Digital watermarking algorithms and watermark attack algorithms evolve in a competitive dynamic, but the lack of progress in watermark attack algorithms is currently limiting the advancement of digital watermarking. This paper addresses this issue by leveraging the strengths of UNet to enhance the Denoising Spatial-Adaptive Network (SADNet), and proposes a Highly Applicable and Imperceptible Watermark Attack Network (HAIWAN). Firstly, contextual modules based on skip connections and the UNet structure are integrated into the network, expanding the receptive field and enhancing the imperceptibility of the output image. Secondly, the residual network is improved, and the Fast and Precise block (FPblock) is introduced to increase training stability. Finally, the connection between image denoising and watermark attacks is explored, with targeted denoising techniques applied to effectively attack watermarks. The experimental results show that the proposed method effectively attacks watermarking algorithms while maintaining image visual quality. Comparisons with other methods further demonstrate the superiority of our method.

关键词： Watermark attack Imperceptibility Convolutional neural Network Digital watermarking algorithm Robustness evaluation

来源：评论

学校读者我要写书评

暂无评论

Deep learning serves traffic safety analysis: A forward-looking review

引用

IET INTELLIGENT TRANSPORT SYSTEMS 2023年第1期17卷 22-71页

作者： Razi, Abolfazl Chen, Xiwen Li, Huayu Wang, Hao Russo, Brendan Chen, Yan Yu, Hongbin Clemson Univ Sch Comp Clemson SC 29634 USA Univ Arizona Dept Elect & Comp Engn Tucson AZ 85721 USA No Arizona Univ Dept Civil Engn Flagstaff AZ 86011 USA Arizona State Univ Polytech Sch Mesa AZ USA Arizona State Univ Sch Elect Comp & Energy Engn Tempe AZ USA

This paper explores deep learning (DL) methods that are used or have the potential to be used for traffic video analysis, emphasising driving safety for both autonomous vehicles and human-operated vehicles. A typical processing pipeline is presented, which can be used to understand and interpret traffic videos by extracting operational safety metrics and providing general hints and guidelines to improve traffic safety. This processing framework includes several steps, including video enhancement, video stabilisation, semantic and incident segmentation, object detection and classification, trajectory extraction, speed estimation, event analysis, modelling, and anomaly detection. The main goal is to guide traffic analysts to develop their own custom-built processing frameworks by selecting the best choices for each step and offering new designs for the lacking modules by providing a comparative analysis of the most successful conventional and DL-based algorithms proposed for each step. Existing open-source tools and public datasets that can help train DL models are also reviewed. To be more specific, exemplary traffic problems are reviewed and required steps are mentioned for each problem. Besides, connections to the closely related research areas of drivers' cognition evaluation, crowd-sourcing-based monitoring systems, edge computing in roadside infrastructures, automated driving systems-equipped vehicles are investigated, and the missing gaps are highlighted. Finally, commercial implementations of traffic monitoring systems, their future outlook, and open problems and remaining challenges for widespread use of such systems are reviewed.

关键词： traffic engineering computing speed estimation traffic analysts operational safety metrics video signal processing remaining challenges learning (artificial intelligence) road safety trajectory extraction providing general hints open problems neural nets video enhancement video stabilisation traffic videos anomaly detection Optical, image and video signal processing Computer vision and image processing techniques traffic safety analysis exemplary traffic problems autonomous vehicles Traffic engineering computing object detection deep learning (artificial intelligence) required steps open-source tools crowd-sourcing-based monitoring systems custom-built processing frameworks driving safety deep learning methods traffic monitoring systems event analysis intelligent transportation systems Video signal processing road traffic successful conventional DL-based algorithms traffic video analysis edge computing road vehicles typical processing pipeline semantic incident segmentation processing framework comparative analysis automated driving systems-equipped vehicles

来源：评论

学校读者我要写书评

暂无评论

NFCNN: toward a noise fusion convolutional neural network for image denoising

引用

signal image AND VIDEO processing 2022年第1期16卷 175-183页

作者： Xu, Maoyuan Xie, Xiaoping Sichuan Univ Sch Math Chengdu 610064 Sichuan Peoples R China

Deep learning-based methods have achieved the state-of-the-art performance in image denoising. In this paper, a deep learning-based denoising method is proposed and a module called fusion block is introduced in the convolutional neural network. For this so-called noise fusion convolutional neural network (NFCNN), there are two branches in its multistage architecture. One branch aims to predict the latent clean image, while the other one predicts the residual image. A fusion block is contained between every two stages by taking the predicted clean image and the predicted residual image as a part of inputs, and it outputs a fused result to the next stage. NFCNN has an attractive texture-preserving ability because of the fusion block. To train NFCNN, a stage-wise supervised training strategy is adopted to avoid the vanishing gradient and exploding gradient problems. Experimental results show that NFCNN is able to perform competitive denoising results when compared with some state-of-the-art algorithms.

关键词： Deep learning Fusion block Two branches network architecture image denoising

来源：评论

学校读者我要写书评

暂无评论

Quantization-Aware Training for Mixed Precision Photonic neural Networks 33

Quantization-Aware Training for Mixed Precision Photonic Neu...

引用

33rd IEEE International Workshop on Machine Learning for signal processing, MLSP 2023

作者： Kirtas, Manos Passalis, Nikolaos Oikonomou, Athina Moralis-Pegios, Miltos Giamourgiannis, George Tsakyridis, Apostolos Mourgias-Alexandris, George Pleros, Nikolaos Tefas, Anastasios Aristotle University of Thessaloniki Computational Intelligence and Deep Learning Research Group Department of Informatics Thessaloniki Greece Aristotle University of Thessaloniki Department of Informatics Thessaloniki Greece Aristotle University of Thessaloniki Wireless and Photonic Systems and Networks Research Group Department of Informatics Thessaloniki Greece

ISBN: (纸本)9798350324112

Neuromorphic photonics is among the most promising approaches for Deep Learning (DL) hardware accelerators with recent state-of-the-art mixed precision configurations lowering even further the energy consumption while increasing simultaneously the computation rate. To this end, we propose a novel stochastic mixed precision quantization-aware training scheme oriented to Photonic neural Networks (PNNs) that gradually lowers the bit resolution of layers based on their position and the observed tolerance to quantization noise. The effectiveness of the proposed method is demonstrated on an image classification task applying a ResNet8 architecture and a financial time series forecasting applying a Recurrent neural Network (RNN) using two photonic configurations. © 2023 IEEE.

关键词： stochastic systems

来源：评论

学校读者我要写书评

暂无评论

ERF-FEM: The Feature Extraction Module based on Effective Receptive Field used for image Inpainting 17

ERF-FEM: The Feature Extraction Module based on Effective Re...

引用

17th IEEE International Conference on signal processing, ICSP 2024

作者： Feng, Yahui Chen, Li Wang, Xiaofeng Zhao, Shulin Chai, Enhui Northwest University School of Information Science and Technology Xi'an China

ISBN: (纸本)9798350387384

Many image inpainting models that rely on convolutional neural networks (CNN) have been extensively utilized in diverse image inpainting contexts, achieving outstanding results. Regrettably, due to the constraints of convolutional kernels, these approaches lack the capacity to capture multi-scale features. On the contrary, the majority of image inpainting methods regarded corrupted images as common pixel consistency images, disregarding the influence of the corrupted regions, such as shifts in mean and variance. To address the aforementioned challenges, we propose an image inpainting model that leverages a multi-scale feature extraction module and attention-enhanced learnable region normalization. It is capable of obtaining feature maps of different scales without stacking multiple layers of convolutions, alleviating the problems of detail information loss and low local computational efficiency. Additionally, during normalization, the learnable region normalization enhanced by attention can mitigate the offset effect of invalid pixels on mean and variance. Our method was verified on datasets Places2 and CelebA, and it was indicated by the experimental results that the similar methods currently accessible were outperformed by our method. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

MG-CNFNet: A multiple grained channel normalized fusion networks for medical image deblurring

引用

BIOMEDICAL signal processing AND CONTROL 2023年第1期82卷 000-000页

作者： Zeng, Xianhua Dong, Qian Li, Yancheng Chongqing Univ Posts & Telecommun Coll Comp Sci & Technol Chongqing 400065 Peoples R China

In different medical imaging techniques, the relative displacement between the patient and detection instru-ment can lead to different degrees of motion blur, which will affect the diagnosis of the patient's condition. In this paper, we propose a deblurring model with Multiple Grained Channel Normalized Fusion Networks (MG-CNFNet), which decomposes the image input by multi-grained division to cascade feature extraction from fine to coarse, constructs the encoder using semi-channel self-normalization to enrich the multi-scale information of the image, highlight the main features and improve the deblurring performance, and uses a shallow convolutional fusion block to obtain more shallow information and balance the performance and efficiency of deblurring. Compared with other state-of-the-art methods, our method is qualitatively and quantitatively evaluated on three medical image datasets (including X-ray, Ultrasound, and MRI) to demonstrate its effectiveness in removing motion blur from medical images, and yields better restoration results.

关键词： Multi grain Semi-channel self-normalization Shallow convolutional fusion Deep neural network Medical image deblurring

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：