检索结果-内蒙古大学图书馆

2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2021

作者： Vinogradova, Nina S. Ivanov, Oleg Yu. Sevastianova, Natalya Yu. Ural Federal University Yekaterinburg Russia

ISBN: (纸本)9781728176918

This work is devoted to the development of an algorithm for recovering unidentified multispectral remote monitoring data. It is noted that during storage, transmission over a distance, preprocessing, Earth remote sensing data may be damaged. The most difficult case of data corruption involves corrupting header information. It is shown that Fourier analysis of the image presented in the row vector format can be used to recover the header information. The analysis of the main regularities of the Fourier transforms of such a signal for the cases of its storage in the BIL, BIP and BSQ formats is carried out. An algorithm for identifying a method for storing Earth remote sensing data is proposed, which involves converting a multispectral image into a one-dimensional digital signal, constructing a spectrum of the received signal, its threshold processing and analyzing the location of the detected peaks. The work of the algorithm is demonstrated by the example of determining the data storage format of the SPOT spacecraft for the case of processing a typical fragment. A study of the functioning of the algorithm when processing heterogeneous images is carried out and recommendations for its use are developed. The results of processing real remote monitoring data obtained by various space-based systems are presented. © 2021 IEEE.

关键词： remote sensing

来源：评论

学校读者我要写书评

暂无评论

Recognizing Human-Object Interaction (HOI) Using Wrist-Mounted Inertial Sensors

引用

IEEE SENSORS JOURNAL 2021年第6期21卷 7899-7907页

作者： Tahir, Samama Raheel, Aasim Ehatisham-Ul-Haq, Muhammad Arsalan, Aamir Univ Engn & Technol Dept Comp Engn Fac Telecommun & Informat Engn Taxila 47050 Pakistan

Detecting and recognizing objects and human-object interactions have great importance in many application areas, such as security and surveillance, robotics, and health monitoring. Over the past few years, many methods have been introduced for activity and object recognition with the help of image and video processing techniques. However, these techniques are very costly and subjected to privacy challenges. Ubiquitous sensing based on wearable sensors (i.e., accelerometer and gyroscope) provides a viable solution to these challenges. Thus, in this paper, we use the wrist-mounted wearable sensors to propose a new method for recognizing human-object interactions, which entails identifying an object, object interactions, and object users. In this aspect, we chose three objects (book, mobile, and laptop) with overall fourteen different interactions performed by 21 users and recorded their data using the wrist-mounted inertial sensors. Various time-domain features are extracted from the data and classified into objects in the first stage. After that, object-dependent interaction and user recognition are carried in the later stage. Three different classifiers are used for recognition experiments (including random forest (RF), J48, and support vector machine (SVM)). The highest F1-scores of 0.872, 0.888, and 0.876 are achieved using RF classifier for recognizing book, mobile, and laptop respectively. The best-case results demonstrate the significance of our proposed method.

关键词： Sensors Wrist Object recognition pattern recognition Gyroscopes Portable computers image sensors Human-object interaction inertial sensor machine learning object recognition sensor fusion wrist sensor

来源：评论

学校读者我要写书评

暂无评论

Muti-Scale Context-Aware Network for Cross-Domain Unsupervised remote sensing image Semantic Segmentation

SSRN

引用

SSRN 2023年

作者： Wang, Xinyao Wang, Haitao Wang, Lingfeng Pan, Chunhong Zhang, Luyang Yan, Liang Li, Jie Huang, Dehai College of Automation Engineering Nanjing University of Aeronautics and Astronautics Nanjing China College of Information Science and Technology Beijing University of Chemical Technology Beijing China National Laboratory of Pattern Recognition Institute of Automation Chinese Academy of Sciences Beijing China Alibaba Group Beijing China China United Network Communications Co. LDT Beijing China Nanjing Huajian Testing Technology Co. LDT Nanjing China

High-resolution remote sensing images contain rich geographical information, and semantic segmentation is widely used as a significant method of scene understanding in remote sensing image processing. However, the convolutional neural network will lose part of the global contextual information in the feature extraction stage. Moreover, due to the different acquisition methods, regions, and times of satellite images, the data distributions and style features of different datasets are also different, which also brings a challenging problem to the cross-domain semantic segmentation task of remote sensing images. To overcome these challenges, a high-resolution context extraction network (CENet) is proposed in this *** first propose a feature fusion module based on spatial and channel attention that can aggregate contextual feature information. In addition, we utilize the multi-level loss functions to constrain the influence of the style differences and unbalanced sample distribution between different datasets. To capture richer contextual information, we design a multi-level feature aggregation to effectively fuse high-level and low-level semantic information. Extensive experiments on two benchmark datasets demonstrate the effectiveness and superiority of our method. The experimental results show that our method achieves better performance compared with other state-of-the-art methods. © 2023, The Authors. All rights reserved.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Radar Fields: An Extension of Radiance Fields to SAR

Radar Fields: An Extension of Radiance Fields to SAR

引用

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

作者： Thibaud Ehret Roger Marí Dawa Derksen Nicolas Gasnier Gabriele Facciolo CNRS ENS Paris-Saclay Centre Borelli Université Paris-Saclay France CNES Toulouse France

ISBN: (数字)9798350365474

ISBN: (纸本)9798350365481

Radiance fields have been a major breakthrough in the field of inverse rendering, novel view synthesis and 3D modeling of complex scenes from multi-view image collections. Since their introduction, it was shown that they could be extended to other modalities such as LiDAR, radio frequencies, X-ray or ultrasound. In this paper, we show that, despite the important difference between optical and synthetic aperture radar (SAR) image formation models, it is possible to extend radiance fields to radar images thus presenting the first "radar fields". This allows us to learn surface models using only collections of radar images, similar to how regular radiance fields are learned and with the same computational complexity on average. Thanks to similarities in how both fields are defined, this work also shows a potential for hybrid methods combining both optical and SAR images.

关键词： Radar remote sensing Laser radar Spaceborne radar Computational modeling Radar imaging Optical imaging Radar polarimetry

来源：评论

学校读者我要写书评

暂无评论

LiDAR-Based Point Cloud Data processing Application Research

LiDAR-Based Point Cloud Data Processing Application Research

引用

Geoscience and remote sensing Mapping (GRSM), International Conference on

作者： Zichuan Zheng Xiaoting Zhang Wanxin Zhou School of Information Engineering Northwest Agriculture and Forestry University Yangling Shaanxi China

LiDAR technology is widely used for point cloud data acquisition in geographic mapping, ecological surveying, etc., which facilitates the research. The PointNet model is a pioneering representative of deep learning technology and semantic segmentation of 3D point cloud, which improves the network generalization ability through a large number of trainings, but at the same time there are problems such as loss of detail data, complex scene recognition accuracy lower, and incomplete local feature extraction. Attention mechanism is a technique in deep learning that enables the network to focus on valid data and discard invalid data in the process of image processing. The attention mechanism can make the local features of point cloud data under deep learning more distinctive, so that the local features can be extracted more completely and accurately. In this paper, we propose to use PointN et network, combined with our own collected dataset for semantic segmentation experiments. And the attention mechanism is introduced to improve the recognition ability of Point Net network as well as the accuracy of segmentation.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Self-Supervised SAR image Denoising Method for Ship Target Detection

Self-Supervised SAR Image Denoising Method for Ship Target D...

引用

IEEE International Conference on Radar

作者： Yun Zhang Weimin Kong Weirui Zhou Haoxuan Yuan School of Electronics and Information Engineering Harbin Institute of Technology China

ISBN: (数字)9798350362381

ISBN: (纸本)9798350362398

In synthetic aperture radar (SAR) images, ship targets are easily submerged by complex background interference, and speckle noise causes blurring of ship edges, severely reducing image quality. This not only affects the usability of the images but also poses challenges for subsequent processing tasks such as image segmentation, target detection, and recognition. To address this issue, this paper proposes a self-supervised SAR image denoising method for ship target detection. This method extracts the regions of interest and suppresses the background in the image using a clustering-based mask. By combining a self-supervised learning strategy, the denoising network is trained only on noisy SAR data without the need for clean reference images, resulting in high-quality images. Experimental verification on GF3 data demonstrates the potential application of the proposed convolutional neural network in SAR image processing.

关键词： Radar remote sensing Noise reduction Noise Object detection Interference Radar imaging Radar polarimetry Marine vehicles Usability Synthetic aperture radar

来源：评论

学校读者我要写书评

暂无评论

4th International Conference on Intelligent Systems and pattern recognition, ISPR 2024

4th International Conference on Intelligent Systems and Patt...

引用

4th International Conference on Intelligent Systems and pattern recognition, ISPR 2024

ISBN: (纸本)9783031821523

The proceedings contain 77 papers. The special focus in this conference is on Intelligent Systems and pattern recognition. The topics include: Tunisian Sign Language recognition and Translation Using Deep Learning;Beyond Spatial: A Wavelet Fusion-Based Deep Learning CAD for Skin Cancer Diagnosis;integrating Machine Learning and Deep Learning Techniques for Enhanced Historical Script Classification;efficiency Meets Resilience: Accelerating Object Detection in Embedded Environments Through Compressive sensing;PMoET: Going Wider Than Deeper Using the Parallel Mixture of Experts Transformer for 3D Hand Gesture recognition;Enhancing Forest Fire Classification with Feature Selection and Machine Learning Based on PRISMA Hyperspectral Data;semantic Segmentation of Unmanned Aerial Vehicle remote sensing images Using SegFormer;handcrafted and Deep Features for Synthetic Data Generation in Offline Handwritten Signature Verification;refining U-Net Architecture Through Genetic Algorithms for Improved Skin Lesion image Segmentation;An RNN-LSTM Approach for Algerian Accent Identification;ensemble Methods for Predicting Cyanobacteria’s Potential Toxicity in Water Dams;cross-Organ Investigation of Tumor Histological Features Similarities Using Transfer Learning: A Case Study on Breast and Colorectal Tumors;behavior-Based Insider Threat Detection Using a Deep Neural Network;holistic Ontology Alignment Using Ontologies Embeddings;learning Different Separations in Branch and Cut: A Survey;Cancer Classification in Breast Imaging via Enhanced CNN Deep Learning Architecture;X-Ray Insights: A Siamese with CNN and Spatial Attention Network for Innovative Person Identification;hybrid Wiener Filter and Sharpening Filter for image De-blurring;feature Extraction and Dimensionality Reduction to Evaluate Decision Tree Ensembles for Diabetic Retinopathy Detection;artificial Intelligence-Based on Automatic Detection of Diabetic Eye Diseases : A Systematic Review;automated Breast Cancer Detection: A

关键词：

来源：评论

学校读者我要写书评

暂无评论

4th International Conference on Intelligent Systems and pattern recognition, ISPR 2024

4th International Conference on Intelligent Systems and Patt...

引用

4th International Conference on Intelligent Systems and pattern recognition, ISPR 2024

ISBN: (纸本)9783031821554

关键词：

来源：评论

学校读者我要写书评

暂无评论

Machine-learned Regularization and Polygonization of Building Segmentation Masks 25

Machine-learned Regularization and Polygonization of Buildin...

引用

25th International Conference on pattern recognition (ICPR)

作者： Zorzi, Stefano Bittner, Ksenia Fraundorfer, Friedrich Graz Univ Technol Inst Comp Graph & Vis Graz Austria German Aerosp Ctr DLR Remote Sensing Technol Inst Cologne Germany

ISBN: (纸本)9781728188089

We propose a machine learning based approach for automatic regularization and polygonization of building segmentation masks. Taking an image as input, we first predict building segmentation maps exploiting generic fully convolutional network (FCN). A generative adversarial network (GAN) is then involved to perform a regularization of building boundaries to make them more realistic, i.e., having more rectilinear outlines which construct right angles if required. This is achieved through the interplay between the discriminator which gives a probability of input image being true and generator that learns from discriminator's response to create more realistic images. Finally, we train the backbone convolutional neural network (CNN) which is adapted to predict sparse outcomes corresponding to building corners out of regularized building segmentation results. Experiments on three building segmentation datasets demonstrate that the proposed method is not only capable of obtaining accurate results, but also of producing visually pleasing building outlines parameterized as polygons.

关键词： Buildings

来源：评论

学校读者我要写书评

暂无评论

3D-PV-Locator: Large-scale detection of rooftop-mounted photovoltaic systems in 3D

引用

APPLIED ENERGY 2022年 310卷 118469-118469页

作者： Mayer, Kevin Rausch, Benjamin Arlt, Marie-Louise Gust, Gunther Wang, Zhecheng Neumann, Dirk Rajagopal, Ram Stanford Univ 473 Via Ortega Stanford CA 94305 USA Ludwig Maximilians Univ Munchen Schackstr 4 D-80539 Munich Germany Univ Freiburg Rempartstr 16 D-79098 Freiburg Germany

While photovoltaic (PV) systems are being installed at an unprecedented rate, it is challenging to keep track of them due to their decentralized character and large number. In this paper, we present the 3D-PV-Locator for large-scale detection of roof-mounted PV systems in three dimensions (3D). The 3D-PV-Locator combines information extracted from aerial images and 3D building data by means of deep neural networks for image classification and segmentation, as well as 3D spatial data processing techniques. It thereby extends existing approaches for the automated detection of PV systems from aerial images by also providing their azimuth and tilt angles. We evaluate the 3D-PV-Locator using a large dataset gathered from the official German PV registry in a real-world study with more than one million buildings. In terms of azimuth and tilt angles, our evaluation shows that the 3D-PV-Locator and the official registry coincide for about two thirds of the observations and are within neighboring classes for 84 and 99 percent of the observations, respectively. In terms of detected PV system capacity, we show that the 3D-PV-Locator clearly outperforms existing approaches. It performs particularly well for the groups of small and medium-sized PV systems (3.6-33.1 percent error reduction) and PV systems tilted beyond 40 (25.6-38.1 percent error reduction). The 3D PV system data generated by the 3D-PV-Locator can inform several practical applications, such as improved forecasting of solar generation, the optimized planning and operation of distribution networks, improved integration of electric vehicles, and others. All datasets and pre-trained models associated with this paper are available online.

关键词： Solar panels Renewable energy image recognition Deep learning Computer vision 3D building data remote sensing Aerial imagery

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：