检索结果-内蒙古大学图书馆

Hyperspectral image Denoising via Tensor Low-Rank Prior and Unsupervised deep Spatial-Spectral Prior

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 2022年 60卷 1页

作者： Wu, Wei-Hao Huang, Ting-Zhu Zhao, Xi-Le Wang, Jian-Li Zheng, Yu-Bang Univ Elect Sci & Technol China Sch Math Sci Res Ctr Image & Vis Comp Chengdu 610056 Peoples R China

Hyperspectral image (HSI) denoising is a fundamental task in remote sensing image processing, which is helpful for HSI subsequent applications, such as unmixing and classification. Thanks to the powerful representation ability of untrained deep neural networks (DNNs), deep image prior (DIP)-based methods achieve tremendous successes in image processing (e.g., denoising and inpainting). However, DIP-based methods neglect the tensor low-rank prior of the underlying HSI, which will be beneficial to capture the global structure of the underlying HSI. To address this issue, we propose a novel model for HSI denoising, which can simultaneously take respective advantages of the tensor low-rank prior and the deep spatial-spectral prior. The tensor low-rank prior leads to a better global structure, and the deep spatial-spectral prior is complementary to preserve better local details. On the one hand, we adopt low-rank tensor ring (TR) decomposition to characterize the tensor low-rank prior and capture the global structure of the underlying HSI. On the other hand, we employ untrained DNNs to flexibly represent the deep spatial-spectral prior and capture the local details of the underlying HSI. To solve the proposed model, we develop an efficient alternating minimization algorithm. Experimental results on simulated and real data validate the advantages of the proposed model in HSI denoising. Compared with state-of-the-art HSI denoising methods, the proposed method preserves better local details and the global structure of the underlying HSI.

关键词： Tensors Noise reduction Correlation learning systems Training data Noise measurement Indexes deep spatial-spectral prior tensor low-rank prior tensor ring (TR) decomposition

来源：评论

学校读者我要写书评

暂无评论

deep learning for Proactive Network Monitoring and Security Protection

引用

IEEE ACCESS 2020年 8卷 19696-19716页

作者： Nguyen Giang Dlugolinsky, Stefan Tran Viet Lopez Garcia, Alvaro Slovak Acad Sci IISAS Inst Informat Bratislava 84507 Slovakia CSIC UC Inst Fis Cantabria IFCA Santander 39005 Spain

The work presented in this paper deals with a proactive network monitoring for security and protection of computing infrastructures. We provide an exploitation of an intelligent module, in the form of a as a machine learning application using deep learning modeling, in order to enhance functionality of intrusion detection system supervising network traffic flows. Currently, intrusion detection systems work well for network monitoring in near real-time and they effectively deal with threats in a reactive way. deep learning is the emerging generation of artificial intelligence techniques and one of the most promising candidates for intelligence integration into traditional solutions leading to quality improvement of the original solutions. The work presented in this paper faces the challenge of cooperation between deep learning techniques and large-scale data processing. The outcomes obtained from extensive and careful experiments show the applicability and feasibility of simultaneously modelled multiple monitoring channels using deep learning techniques. The proper joining of deep learning modelling with scalable data preprocessing ensures high quality and stability of model performance in dynamic and fast-changing environments such as network traffic flow monitoring.

关键词： deep learning proactive forecasting network monitoring cyber security anomaly detection neural machine translation

来源：评论

学校读者我要写书评

暂无评论

Road type classification using deep learning for Tire-Pavement interaction noise data in autonomous driving vehicle

引用

APPLIED ACOUSTICS 2023年第1期212卷

作者： Lee, Sang -Kwon Yoo, Jinhwan Lee, Chang-Hun An, Kanghyun Yoon, Young -Sam Lee, Jaehun Yeom, Gi-Ho Hwang, Seoung-Uk Inha Univ Acoust & Vibrat Signal Proc Lab Incheon South Korea Hyundai Motor Co Seoul South Korea NEXEN Tire Corp Yangsan South Korea

This study presents a novel artificial intelligence model (AIM) for the real-time classification of 13 different road types in an autonomous vehicle. The model was developed based on a combination of a continuous wavelet transform (CWT) and convolutional neural network (CNN). Previously, three methods have been used for the classification of road types, depending on the type of sensor. First, a camera sensor has been widely used because it can capture the road type directly. Second, a vibration sensor has been used, since the vibration level measured on the suspension or inside the tire depends on the road type. Finally, an acoustic sensor has been used, especially in measuring tire-pavement interaction noise (TPIN). In a previous feasibility study, an AIM was developed to classify road types using TPIN signals, which vary depending on the road type. It can distinguish between two road types: asphalt and snow. Recently, CNN has been widely used as an AIM for classification, but it is limited as the input size of the CNN should be optimized for real-time processing due to its long calculation times, even for a 2D convolution process. Its input is image data, which can be produced through the CWT of the TPIN signal. This study proposes an AIM that can classify 13 different road surfaces in real-time while driving. In this study, a method to determine the optimal filter band and data length used for CWT is proposed. The method was developed based on the classification accuracy of an AIM. The developed AIM was successfully applied to the real-time classification of road types with an accuracy of 95% on a public road.

关键词： Continuous wavelet transform Convolution neural network Tire -pavement interaction noise Classification Road surfaces

来源：评论

学校读者我要写书评

暂无评论

An Efficient Information-Reinforced Lidar deep Completion Network without RGB Guided

引用

REMOTE SENSING 2022年第19期14卷 4689-4689页

作者： Wei, Ming Zhu, Ming Zhang, Yaoyuan Sun, Jiaqi Wang, Jiarong Chinese Acad Sci Changchun Inst Opt Fine Mech & Phys Changchun 130033 Peoples R China Univ Chinese Acad Sci Beijing 100049 Peoples R China

Due to the sparsity of point clouds obtained by LIDAR, the depth information is usually not complete and dense. The depth completion task is to recover dense depth information from sparse depth information. However, most of the current deep completion networks use RGB images as guidance, which are more like a processing method of information fusion. They are not valid when there is only sparse depth data and no other color information. Therefore, this paper proposes an information-reinforced completion network for a single sparse depth input. We use a multi-resolution dense progressive fusion structure to maximize the multi-scale information and optimize the global situation by point folding. At the same time, we re-aggregate the confidence and impose another depth constraint on the pixel depth to make the depth estimation closer to the ground trues. Our experimental results on KITTI and NYU Depth v2 datasets show that the proposed network achieves better results than other unguided deep completion methods. And it is excellent in both accuracy and real-time performance.

关键词： depth completion lidar data processing image processing deep learning

来源：评论

学校读者我要写书评

暂无评论

DIFFRACTION WAVE SEPARATION AND IMAGING WITH deep learning NETWORK: APPLICATION OF GPR DATA 82

DIFFRACTION WAVE SEPARATION AND IMAGING WITH DEEP LEARNING N...

引用

82nd EAGE Conference and Exhibition 2021

作者： Ma, M. Guo, Y. Chang'an University China University of Louisiana Lafayette United States

ISBN: (纸本)9781713841449

We propose a technique of diffraction separation and imaging based on the deep learning network. In order to image the small scale anomalous objects, the diffraction wavefield is identified and extracted from the conventional zero-offset profile (GPR or seismic stacked data) with our trained network. The migration velocity analysis with unique diffraction wavefield and migration in time domain are deployed in the next processing. Finally, the updated data has a better spatial and time resolution which is compared with original profile. Although in the part of neural network training only the classified synthetic records are generated and built as the datasets, the performance of separation work is still high and next diffraction imaging can be completed successfully, which is verified with the GPR data test in the real application area. © (2021) by the European Association of Geoscientists & Engineers (EAGE)

关键词： Diffraction

来源：评论

学校读者我要写书评

暂无评论

A Bayesian clustering ensemble Gaussian process model for network-wide traffic flow clustering and prediction

引用

TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES 2023年第1期148卷

作者： Zhu, Zheng Xu, Meng Ke, Jintao Yang, Hai Chen, Xiqun (Michael) Zhejiang Univ Inst Intelligent Transportat Syst Coll Civil Engn & Architecture Hangzhou Peoples R China Alibaba Zhejiang Univ Joint Res Inst Frontier Tech Hangzhou Peoples R China Hong Kong Univ Sci & Technol Dept Civil & Environm Engn Hong Kong Peoples R China Univ Hong Kong Dept Civil Engn Hong Kong Peoples R China Hong Kong Univ Sci & Technol Guangzhou Intelligent Transportat Thrust Guangzhou Peoples R China Zhejiang Univ Haining Peoples R China Univ Illinois Urbana Champaign Inst Champaign IL 61820 USA Zhejiang Prov Engn Res Ctr Intelligent Transportat Hangzhou Peoples R China

Traffic flow prediction is an essential component in intelligent transportation systems. Recently, there has been a notable trend in applying machine learning models, especially deep learning, for network-wide traffic prediction. However, existing studies have limitations on model interpretability, model generalization, and over-reliance on image data processing or fine-designed deep learning structures for extracting traffic attributes. This paper attempts to tackle these limitations by proposing a Bayesian clustering ensemble Gaussian process (BCEGP) model for network-wide traffic flow clustering and prediction. The model utilizes a subset-based Dirichlet process mixture (SDPM) model to conduct a hard clustering among input data;then, within each cluster, it adopts the Gaussian Process (GP) to learn the probability relationship between inputs and outputs. During the prediction phase, the model conducts a soft clustering of the input as weights, and makes predictions via a weighted average of GPs' outputs. The merits of the BCEGP model include: (a) data with similar spatial-temporal patterns are clustered, which helps understand traffic dynamics in a non-Euclidean and non-graphical manner that enhances information extracting for model development;(b) GPs provide analytically trackable functions/gradients of predicted traffic flows with features and reveal variances of predicted traffic flow, enhancing model applicability and interpretability to some extent;(c) the model incorporates an ensemble learning framework that achieves great generalization performance as good as deep learning models;(d) the subset-based clustering and cluster-based GP learning are conducted parallelly, and thus vastly accelerate the training efficiency compared with conventional GPs (but slower than deep learning models). We test the performance of the proposed model based on both synthesized and real-world datasets. For comparison, several widely used machine learning and deep learning mod

关键词： Gaussian Process Statistical learning Traffic flow prediction Dirichlet process mixture model

来源：评论

学校读者我要写书评

暂无评论

Dataset Condensation via Expert Subspace Projection

引用

SENSORS 2023年第19期23卷 8148-8148页

作者： Ma, Zhiheng Gao, Dezheng Yang, Shaolei Wei, Xing Gong, Yihong Chinese Acad Sci Shenzhen Inst Adv Technol Shenzhen 518055 Peoples R China Xi An Jiao Tong Univ Inst Artificial Intelligence & Robot Xian 710049 Peoples R China Jiaotong Univ Sch Software Engn Xian 710049 Peoples R China

The rapid growth in dataset sizes in modern deep learning has significantly increased data storage costs. Furthermore, the training and time costs for deep neural networks are generally proportional to the dataset size. Therefore, reducing the dataset size while maintaining model performance is an urgent research problem that needs to be addressed. Dataset condensation is a technique that aims to distill the original dataset into a much smaller synthetic dataset while maintaining downstream training performance on any agnostic neural network. Previous work has demonstrated that matching the training trajectory between the synthetic dataset and the original dataset is more effective than matching the instantaneous gradient, as it incorporates long-range information. Despite the effectiveness of trajectory matching, it suffers from complex gradient unrolling across iterations, which leads to significant memory and computation overhead. To address this issue, this paper proposes a novel approach called Expert Subspace Projection (ESP), which leverages long-range information while avoiding gradient unrolling. Instead of strictly enforcing the synthetic dataset's training trajectory to mimic that of the real dataset, ESP only constrains it to lie within the subspace spanned by the training trajectory of the real dataset. The memory-saving advantage offered by our method facilitates unbiased training on the complete set of synthetic images and seamless integration with other dataset condensation techniques. Through extensive experiments, we have demonstrated the effectiveness of our approach. Our method outperforms the trajectory matching method on CIFAR10 by 16.7% in the setting of 1 image/Class, surpassing the previous state-of-the-art method by 3.2%.

关键词： dataset condensation synthetic data subspace optimization deep learning

来源：评论

学校读者我要写书评

暂无评论

Self‐Supervised learning for Solar Radio Spectrum Classification

arXiv

引用

arXiv 2025年

作者： Li, Siqi Yuan, Guowu Chen, Jian Tan, Chengming Zhou, Hao School of Information Science & Engineering Yunnan University Kunming650500 China CAS Key Laboratory of Solar Activity National Astronomical Observatories Beijing100012 China State Key Laboratory of Space Weather National Space Science Center Beijing100864 China School of Astronomy and Space Sciences University of Chinese Academy of Sciences Beijing100049 China

Solar radio observation is an important way to study the Sun. Solar radio bursts contain important information about solar activity. Therefore, real‐time automatic detection and classification of solar radio bursts are of great value for subsequent solar physics research and space weather warnings. Traditional image classification methods based on deep learning often require considerable training data. To address insufficient solar radio spectrum images, transfer learning is generally used. However, the large difference between natural images and solar spectrum images has a large impact on the transfer learning effect. In this paper, we propose a self‐supervised learning method for solar radio spectrum classification. Our method uses self‐supervised training with a self‐masking approach in natural language processing. Self‐supervised learning is more conducive to learning the essential information about images compared with supervised methods, and it is more suitable for transfer learning. First, the method pre‐trains using a large amount of other existing data. Then, the trained model is fine‐tuned on the solar radio spectrum dataset. Experiments show that the method achieves a classification accuracy similar to that of convolutional neural networks and Transformer networks with supervised training. © 2025, CC BY-NC-ND.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

image real-time augmented reality technology based on spatial color and depth consistency

Image real-time augmented reality technology based on spatia...

引用

作者： Zhai, Lijie Chen, Donghui Department of Physics and Electrical Engineering Weinan Normal University WeinanShanxi714099 China National Weather Information Center Beijing100081 China

Augmented reality can enhance people’s perception of the environment by embedding virtual objects or other information in real-time images. In this paper, the color image is used as a reference to calculate the confidence of the original depth map, and stereo matching is performed according to the feature points. The depth map is mainly enhanced by the color, edge, and segmentation results of the color image. A deep computing system based on augmented reality is designed. The system can use a binocular camera to collect object images in real time and obtain better parallax images by correcting the calibrated image. The semi-global block matching algorithm and depth calculation are used to realize the tracking registration of virtual objects. Experiments in different environments show that the system has good real-time performance, light invariance, and depth consistency. © 2020, Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Augmented reality

来源：评论

学校读者我要写书评

暂无评论

Advanced Underwater image Quality Enhancement via Hybrid Super-Resolution Convolutional Neural Networks and Multi-Scale Retinex-Based Defogging Techniques

arXiv

引用

arXiv 2024年

作者： Gogireddy, Yugandhar Reddy Gogireddy, Jithendra Reddy Dept. Computational Intelligence SRM Institute of Science and Technology Chennai India Dept. Computer Science and Engineering SRM University AP India

The difficulties of underwater image degradation due to light scattering, absorption, and fog-like particles which lead to low resolution and poor visibility are discussed in this study report. We suggest a sophisticated hybrid strategy that combines Multi-Scale Retinex (MSR) defogging methods with Super-Resolution Convolutional Neural Networks (SRCNN) to address these problems. The Retinex algorithm mimics human visual perception to reduce uneven lighting and fogging, while the SRCNN component improves the spatial resolution of underwater photos.[11] Through the combination of these methods, we are able to enhance the clarity, contrast, and colour restoration of underwater images, offering a reliable way to improve image quality in difficult underwater conditions. The research conducts extensive experiments on real-world underwater datasets to further illustrate the efficacy of the suggested approach. In terms of sharpness, visibility, and feature retention, quantitative evaluation which use metrics like the Structural Similarity Index Measure (SSIM) and Peak Signal-to-Noise Ratio (PSNR) demonstrates notable advances over conventional techniques.[41] In real-time underwater applications like marine exploration, underwater robotics, and autonomous underwater vehicles, where clear and high-resolution imaging is crucial for operational success, the combination of deep learning and conventional image processing techniques offers a computationally efficient framework with superior results. Copyright © 2024, The Authors. All rights reserved.

关键词： Underwater imaging

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：