检索结果-内蒙古大学图书馆

Dual-Weighted Kernel Extreme Learning Machine for Hyperspectral imagery Classification

remote sensing 2021年第3期13卷 1-21页

作者： Yu, Xumin Feng, Yan Gao, Yanlong Jia, Yingbiao Mei, Shaohui Northwestern Polytech Univ Sch Elect & Informat Engn Xian 710072 Peoples R China Shaoguan Univ Sch Elect & Informat Engn Shaoguan 512023 Peoples R China

Due to its excellent performance in high-dimensional space, the kernel extreme learning machine has been widely used in pattern recognition and machine learning fields. In this paper, we propose a dual-weighted kernel extreme learning machine for hyperspectral imagery classification. First, diverse spatial features are extracted by guided filtering. Then, the spatial features and spectral features are composited by a weighted kernel summation form. Finally, the weighted extreme learning machine is employed for the hyperspectral imagery classification task. This dual-weighted framework guarantees that the subtle spatial features are extracted, while the importance of minority samples is emphasized. Experiments carried on three public data sets demonstrate that the proposed dual-weighted kernel extreme learning machine (DW-KELM) performs better than other kernel methods, in terms of accuracy of classification, and can achieve satisfactory results.

关键词： weighted kernel extreme learning dual-weighted multiple scales guided filter hyperspectral image classification imbalanced dataset

来源：评论

学校读者我要写书评

暂无评论

In the light of feature distributions: moment matching for Neural Style Transfer

In the light of feature distributions: moment matching for N...

引用

IEEE/CVF Conference on Computer Vision and pattern recognition (CVPR)

作者： Kalischek, Nikolai Wegner, Jan D. Schindler, Konrad Swiss Fed Inst Technol Photogrammetry & Remote Sensing EcoVis Lab Zurich Switzerland

ISBN: (纸本)9781665445092

Style transfer aims to render the content of a given image in the graphical/artistic style of another image. The fundamental concept underlying Neural Style Transfer (NST) is to interpret style as a distribution in the feature space of a Convolutional Neural Network, such that a desired style can be achieved by matching its feature distribution. We show that most current implementations of that concept have important theoretical and practical limitations, as they only partially align the feature distributions. We propose a novel approach that matches the distributions more precisely, thus reproducing the desired style more faithfully, while still being computationally efficient. Specifically, we adapt the dual form of Central Moment Discrepancy (CMD), as recently proposed for domain adaptation, to minimize the difference between the target style and the feature distribution of the output image. The dual interpretation of this metric explicitly matches all higher-order centralized moments and is therefore a natural extension of existing NST methods that only take into account the first and second moments. Our experiments confirm that the strong theoretical properties also translate to visually better style transfer, and better disentangle style from semantic image content.

关键词： Measurement Computer vision Semantics Computational efficiency Convolutional neural networks pattern matching

来源：评论

学校读者我要写书评

暂无评论

Linear-Geometry Distortion Correction for Bistatic Inverse Synthetic Aperture Radar Imaging Based on Deep Learning Model

Linear-Geometry Distortion Correction for Bistatic Inverse S...

引用

IEEE International Symposium on Geoscience and remote sensing (IGARSS)

作者： Jiannan Wang Jingtao Ma Penghui Huang Shengqi Zhu Haojuan Yuan Yanyang Liu Anjie Cao Xiangcheng Wan Muyang Zhan School of Electronic Information and Engineering Shanghai Jiao Tong University Shanghai China School of Electrical and Electronic Engineering Shanghai Institute of Technology Shanghai China National Laboratory of Radar Signal Processing Xidian University Xi’an China Shanghai Aerospace Electronics Technology Institute Shanghai China Shanghai Institute of Satellite Engineering Shanghai China

ISBN: (数字)9798350360325

ISBN: (纸本)9798350360332

Compared with a monostatic inverse synthetic aperture radar (ISAR) imaging system, a bistatic ISAR (Bi-ISAR) system offers more comprehensive target information, along with higher system security and resistance to interference. However, the inherent geometric configuration of Bi-ISAR will introduce the linear-geometry distortion (LGD), causing the target to appear sheared in shape and impacting subsequent target recognition. To deal with this issue, in this paper, a novel deep learning-based algorithm is proposed to realize the LGD correction. In the proposed algorithm, a neural network called DoubleUNet is employed for the accurate semantic segmentation of target's ISAR image. Then the initial target ISAR image is transformed into two-dimensional point clouds based on the estimated radar parameters. Finally, the linear coupling relationship between azimuth and range dimensions is removed, beneficial to the target ISAR shape restoration and subsequent recognition. Simulation experiments validate the proposed algorithm.

关键词： Deep learning Target recognition Shape Simulation Semantic segmentation Imaging Radar imaging

来源：评论

学校读者我要写书评

暂无评论

Enrich, Distill and Fuse: Generalized Few-Shot Semantic Segmentation in remote sensing Leveraging Foundation Model’s Assistance

Enrich, Distill and Fuse: Generalized Few-Shot Semantic Segm...

引用

IEEE Computer Society Conference on Computer Vision and pattern recognition Workshops (CVPRW)

作者： Tianyi Gao Wei Ao Xing-ao Wang Yuanhao Zhao Ping Ma Mengjie Xie Hang Fu Jinchang Ren Zhi Gao Wuhan University Robert Gorden University

ISBN: (数字)9798350365474

ISBN: (纸本)9798350365481

Generalized few-shot semantic segmentation (GFSS) unifies semantic segmentation with few-shot learning, showing great potential for Earth observation tasks under data scarcity conditions, such as disaster response, urban planning, and natural resource management. GFSS requires simultaneous prediction for both base and novel classes, with the challenge lying in balancing the segmentation performance of both. Therefore, this paper introduces a novel framework named FoMA, Foundation Model Assisted GFSS framework for remote sensing images. We aim to leverage the generic semantic knowledge inherited in foundation models. Specifically, we employ three strategies named Support Label Enrichment (SLE), Distillation of General Knowledge (DGK) and Voting Fusion of Experts (VFE). For the support images, SLE explores credible unlabeled novel categories, ensuring that each support label contains multiple novel classes. For the query images, DGK technique allows an effective transfer of generalizable knowledge of foundation models on certain categories to the GFSS learner. Additionally, VFE strategy integrates the zero-shot prediction of foundation models with the few-shot prediction of GFSS learners, achieving improved segmentation performance. Extensive experiments and ablation studies conducted on the OpenEarthMap few-shot challenge dataset demonstrate that our proposed method achieves state-of-the-art performance.

关键词： Natural resources image recognition Annotations Fuses Semantic segmentation Urban planning Semantics

来源：评论

学校读者我要写书评

暂无评论

SPATIAL pattern ANALYSIS THROUGH DISTRIBUTION METRICS 24

SPATIAL PATTERN ANALYSIS THROUGH DISTRIBUTION METRICS

引用

24th ISPRS Congress on Imaging Today, Foreseeing Tomorrow

作者： Biraghi, C. A. Lenzi, E. Politecn Milan Dept Architecture Built Environm & Construct Engn Via Giuseppe Ponzio 31 I-20133 Milan Italy Politecn Milan Dipartimento Elettron Informaz & Bioingn Via Giuseppe Ponzio 34-5 I-20133 Milan Italy

Moving from the controversial results on the link between urban structure and performance aspect, this article wants to encourage the development of the independent research on urban structure, and more generally on spatial patterns, at different scales to enable future further correlations with a wider set of performance aspects (environmental, social, economic, medical). The work also exploits the potential of several unsupervised learning algorithms, whose performance and power are increasingly promising and whose use is becoming more widespread in different fields;but for which there are still many challenges concerning the correct application in urban areas and the interpretability of the results. We propose an approach for the creation of new spatial attributes and metrics (features) aiming to quantitatively describe the qualitative distribution of objects (e.g., buildings) in a 2D space. It explores an incremental bottom-up process for the creation of groups of objects (e.g., urban patches) and the evaluation of their physical properties alone and in respect with a sample area at each iteration. The process consists of 7 phases: data preparation, data processing, parameters collection, feature calculation, feature selection, clustering, results comparison. The results can be mainly divided in two. First, the feature selection allowed to extract a minimum set of non-redundant, valid, and consistent features that can explain qualitative distribution aspects of spatial patterns. Second, the comparison between feature-based and neural network clustering, gave useful insights for a preliminary understanding of unsupervised learning techniques internal mechanisms.

关键词： Spatial patterns image analysis Multi-Metrics Feature selection Clustering Neural networks Urban morphology

来源：评论

学校读者我要写书评

暂无评论

CycleGAN-Based SAR-Optical image Fusion for Target recognition

引用

remote sensing 2023年第23期15卷 5569-5569页

作者： Sun, Yuchuang Yan, Kaijia Li, Wangzhe Chinese Acad Sci Aerosp Informat Res Inst Natl Key Lab Microwave Imaging Technol Beijing 100190 Peoples R China Univ Chinese Acad Sci Sch Elect Elect & Commun Engn Beijing 100049 Peoples R China

The efficiency and accuracy of target recognition in synthetic aperture radar (SAR) imagery have seen significant progress lately, stemming from the encouraging advancements of automatic target recognition (ATR) technology based on deep learning. However, the development of a deep learning-based SAR ATR algorithm still faces two critical challenges: the difficulty of feature extraction caused by the unique nature of SAR imagery and the scarcity of datasets caused by the high acquisition cost. Due to its desirable image nature and extremely low acquisition cost, the simulated optical target imagery obtained through computer simulation is considered a valuable complement to SAR imagery. In this study, a CycleGAN-based SAR and simulated optical image fusion network (SOIF-CycleGAN) is designed and demonstrated to mitigate the adverse effects of both challenges simultaneously through SAR-optical image bidirectional translation. SAR-to-optical (S2O) image translation produces artificial optical images that are high-quality and rich in details, which are used as supplementary information for SAR images to assist ATR. Conversely, optical-to-SAR (O2S) image translation generates pattern-rich artificial SAR images and provides additional training data for SAR ATR algorithms. Meanwhile, a new dataset of SAR-optical image pairs containing eight different types of aircraft has been created for training and testing SOIF-CycleGAN. By combining image-quality assessment (IQA) methods and human vision, the evaluation verified that the proposed network possesses exceptional bidirectional translation capability. Finally, the results of the S2O and O2S image translations are simultaneously integrated into a SAR ATR network, resulting in an overall accuracy improvement of 6.33%. This demonstrates the effectiveness of SAR-optical image fusion in enhancing the performance of SAR ATR.

关键词： synthetic aperture radar (SAR) automatic target recognition (ATR) image fusion deep learning CycleGAN

来源：评论

学校读者我要写书评

暂无评论

深度学习在遥感图像处理中的应用

引用

移动信息 2024年第7期46卷 401-403页

作者：郑智闫业国郑州智能科技职业学院郑州450000

随着深度学习的不断发展,其应用领域与应用范围也在不断拓展,对人类的生产生活产生了重要影响,并有力推动了遥感图像处理技术的发展,对于提高遥感影像识别和分类精度具有重要意义。遥感图像处理的要求较高,单纯依靠人工处理遥感图像的... 详细信息

随着深度学习的不断发展,其应用领域与应用范围也在不断拓展,对人类的生产生活产生了重要影响,并有力推动了遥感图像处理技术的发展,对于提高遥感影像识别和分类精度具有重要意义。遥感图像处理的要求较高,单纯依靠人工处理遥感图像的难度较大。深度学习为遥感图像处理提供了新的思路与方法,解决了遥感识别与分类的问题。基于此,文中分析了遥感图像处理技术现状,探究了深度学习在遥感图像处理中的应用。

关键词：深度学习遥感图像处理 SAR图像目标识别神经网络

来源：评论

学校读者我要写书评

暂无评论

First Dataset of Wind Turbine Data Created at National Level With Deep Learning Techniques From Aerial Orthophotographs With a Spatial Resolution of 0.5 M/Pixel

引用

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND remote sensing 2021年 14卷 7968-7980页

作者： Manso-Callejo, Miguel-Angel Cira, Calimanut-Ionut Garrido, Ramon Pablo Alcarria Matesanz, Francisco Javier Gonzalez Univ Politecn Madrid Dept Ingn Topog & Cartog Madrid 28031 Spain Direcc Gen Inst Geog Nacl Subdirecc Gen Geodesia & Cartog Madrid 28003 Spain

Deep learning applied to feature extraction and mapping from high-resolution images is demonstrating the potential of this branch of data-intensive Artificial Intelligence to improve terrain mapping processes. The documented experiences have been applied on a small scale and there is a great expectation about its applicability on a country scale. For example, when extracting wind turbines using semantic segmentation models from a region of 28 km x 19 km containing unseen data, we obtained a commission rate of 1.4% and an omission rate of 0.38%. In this article, we present a methodology generated on the basis of two iterations. In these iterations, processing and postprocessing time, energy consumption, and finally results have been optimized to map wind turbines for the first time throughout the Spanish peninsular territory. In addition to adding a binary classification neural network prior to the semantic segmentation that extracts the turbines, a third multiclass recognition network has been used to classify the turbines by their power capacity complementing the features extracted with attributes. The proposed methodology can be adapted in the vectorization phase and applied to other types of features with linear or polygon representation to achieve a large-scale efficient extraction of geospatial elements using automated procedures.

关键词： Wind turbines Feature extraction image segmentation Training Semantics Task analysis Production Feature extraction feature recognition image classification semantic segmentation wind turbines

来源：评论

学校读者我要写书评

暂无评论

2nd International Conference on 3D Imaging Technologies-Multidimensional Signal processing and Deep Learning, 3DIT-MSPandDL 2020

2nd International Conference on 3D Imaging Technologies-Mult...

引用

2nd International Conference on 3D Imaging Technologies-Multidimensional Signal processing and Deep Learning, 3DIT-MSPandDL 2020

ISBN: (纸本)9789811633904

The proceedings contain 39 papers. The special focus in this conference is on 3D Imaging Technologies-Multidimensional Signal processing and Deep Learning. The topics include: Appearance Defect Detection of Injection Parts Based on Deep Learning;thresholding Method for the Sonar image of a Seabed Target Based on Two-Dimensional Renyi's Entropy;noise Reduction in Sonar images of Seabed Targets Based on a Variational Method;sonar image Segmentation of Seabed Targets Using a Variational Approach;balance the Robustness and Invisibility of Digital Watermarking by image Entropy in Multiple Domains;research Progress of Electrical Resistance Tomography;research on Low-Resolution image Matching and recognition Algorithm Based on Unified Feature Space;research on Realistic Representation Technology in Virtual Environment;intelligent Customer Service Operation Management System Solution Based on Intelligent Voice Analysis;Research Status of Motor imagery EEG Signal Based on Deep Learning;research on Car Seat Comfort Evaluation System Based on Chinese Population;research on Building Extraction from High-Resolution remote sensing image Based on Improved U-Net;a Survey of image Classification Algorithms Based on Graph Neural Networks;an Improved Method of Non-local Means Denoising Based on Histogram;short-Term Traffic Volume Prediction Based on Gray Wolf Optimization Algorithm;Biological Particle recognition Based on BP Neural Network Algorithm;imageCube: A Low-Cost 6D Controller with Smooth Tracking for Mobile Augmented Reality;A Clinical Decision-Making System for COVID-19;License Plate recognition System Based on OpenCV;LUCC and Landscape pattern Variation in the Lower Tarim River by remote sensing;image recognition Based on Super-Resolution Wasserstein Generative Adversarial Nets with Gradient Penalty.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Relation Aware Network for Multi-View SAR Target recognition

Relation Aware Network for Multi-View SAR Target Recognition

引用

Signal, Information and Data processing (ICSIDP), IEEE International Conference on

作者： Yijun Tang Ling Wang Yong Li Daiyin Zhu Key Laboratory of Radar Imaging and Microwave Photonics Ministry of Education College of Electronic and Information Engineering Nanjing University of Aeronautics and Astronautics Nanjing China

ISBN: (数字)9798331515669

ISBN: (纸本)9798331515676

Synthetic Aperture Radar (SAR) has unique advantages and plays an important role in the field of remote sensing, which makes the research of SAR Automatic Target recognition (ATR) very hot. In recent years, due to the rapid development of deep learning networks and hardware computing power, the deep learning based ATR algorithm for SAR has gradually become the mainstream. Compared with the single-view based recognition, multi-view SAR target recognition tends to better utilize the target scattering information of different azimuth angles. In this paper, we propose a Multi-View Relation Aware (MVRA) network for SAR ATR, which can capture the latent relationships between multiple input views, achieving better recognition accuracy than single-view recognition. Additionally, to enable the network to more accurately capture the contour information of SAR image targets, the Laplacian pyramid is applied for SAR image preprocessing. The proposed MVRA network shows a significant improvement in recognition accuracy on the public dataset MSTAR compared to other multi-view SAR target recognition algorithms. Furthermore, on the new Ground Military Vehicle Target (GMVT) dataset constructed by Nanjing University of Aeronautics and Astronautics (NUAA), the proposed MVRA network demonstrates excellent generalization performance even in the scenarios with relatively strong background clutter interference.

关键词： Deep learning Accuracy Laplace equations image recognition Target recognition Feature extraction Radar polarimetry Clutter Synthetic aperture radar Testing

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：