检索结果-内蒙古大学图书馆

2024 International Conference on Intelligent Education and Computer Technology, IECT 2024

作者： Yang, Yang Zhang, Hongyue College of Geography and Oceanography Minjiang University Fujian Fuzhou China Key Lab of Spatial Data Mining and Information Sharing of Ministry of Education Fuzhou University Fujian Fuzhou China

ISBN: (纸本)9798400709920

Large number of publications have been published in marine remote sensing study, and it is of great importance to explore and discover the key research topics and hotspots for domain researchers. In this study, a combination of topic modeling and keywords co-occurrence analysis was applied to 7,509 articles published between 2018 and 2022 obtained from the Web of Science database. The findings of this research indicate that marine remote sensing can be categorized into five major themes: Land Surface Temperature (LST), Sea Surface Temperature (SST), SAR/InSAR for target identification and detection, SAR, and altimetry technology for monitoring oceanographic and atmospheric phenomena, and water quality and organic substance monitoring. Changes in topic strengths reflect a growing emphasis on SAR-based target identification and detection research. keyword co-occurrence evolution results highlight the increasing importance of SAR and Sentinel theme clusters. In terms of research trends, applications in marine remote sensing are likely to prioritize feature extraction, sea measurement, and marine vehicles. In addition to traditional data sources such as MODIS and Landsat, Sentinel, Cryosat-2, AMSR-E and TerraSAR-X data offer significant potential for marine remote sensing applications. © 2024 ACM.

关键词： Surface temperature

来源：评论

学校读者我要写书评

暂无评论

SLK-SLAM: Integrating semantics and optical flow for SLAM in dynamic environments

SLK-SLAM: Integrating semantics and optical flow for SLAM in...

引用

Computer, Control and Robotics (ICCCR), International Conference on

作者： Gang Wang Yu Sun Xin Yu Zhanchao Huang Key Lab of Spatial Data Mining and Information Sharing of Ministry of Education Fuzhou China

ISBN: (数字)9798350373141

ISBN: (纸本)9798350373158

Many contemporary visual SLAM algorithms en-counter significant biases when localizing and mapping in dynamic environments, largely stemming from their extensive reliance on static environment assumptions. In this paper, we propose a semantic SLAM system, called SLK-SLAM, that skill-fully recognizes dynamic objects and generates clean point cloud maps. We determine dynamic objects based on reprojection error, and to improve the number of feature matches in low texture target regions, optical flow is used to track the current frame and tracking sensitivity is applied to measure the tracking error, and ultimately the dynamism of the object is determined with the help of chi-square test. The proposed system is evaluated on the TUM dynamic benchmark dataset. Experimental results show that SLK-SLAM reduces the average absolute tracking error by about 85% compared to ORB-SLAM3. In addition, SLK-SLAM has higher localization accuracy and system robustness compared to other state-of-the-art dynamic SLAM systems.

关键词： Point cloud compression Location awareness Simultaneous localization and mapping Target tracking Accuracy Heuristic algorithms Dynamics

来源：评论

学校读者我要写书评

暂无评论

Unsupervised part extraction of substation equipment based on joint multi-level voxels’ features of point clouds 5

Unsupervised part extraction of substation equipment based o...

引用

5th International Conference on Signal Processing and Computer Science, SPCS 2024

作者： Zhang, Meng Lin, Ding Ding, Jing Fang, Tao Key Lab of Spatial Data Mining and Information Sharing Ministry of Education Fuzhou University Fujian Fuzhou350108 China State Grid Hangzhou Power Supply Company State Grid ZHEJIANG Electric Power co. LTD Zhejiang Hangzhou310000 China

ISBN: (数字)9781510686731

ISBN: (纸本)9781510686724

Aiming at the segmentation and extraction of the main part of substation equipment, we use Fast Point Feature Histograms (FPFH) and Locally Convex Connected Patches (LCCP) to obtain voxels’ integrated geometric features, then aggregate these features and their K nearest neighbors’ on voxels to build multi-level voxels’ features by bottom-up hierarchy, and achieve pre-segmentation of shapes with the flow-constrained super-voxel clustering algorithm;After the pre-segmentation, we conduct shape analysis to extract semantically meaningful instances of equipment components, achieving part-level point cloud data instance extraction of artificial equipment geometric features. Without training data or manual annotations, the work presented is simple and easy to implement. It can merge patches across surface-singularities. It needs a few parameters, can achieve automatic 3D instance extraction from point clouds for different scenes with the same or similar parameters. © 2025 SPIE.

关键词： Superpixels

来源：评论

学校读者我要写书评

暂无评论

Building Type Classification Using Dual-Encoder Interaction Network from Very-High-Resolution Remote Sensing Images

Building Type Classification Using Dual-Encoder Interaction ...

引用

Artificial Intelligence Technology (ACAIT), Asian Conference on

作者： Shaofeng Zhang Mengmeng Li Kangkai Lou Key Lab of Spatial Data Mining & Information Sharing of Ministry of Education Academy of Digital China (Fujian) Fuzhou University FuZhou China

ISBN: (数字)9798331517090

ISBN: (纸本)9798331517106

Building type information is crucial as it indicates the functional properties of structures. The accurate type classification of buildings holds significant importance in smart city development and estimation of carbon emissions. This research explores a novel framework, i.e., hierarchical feature interactive network (HFINet) based on dual-encoder architecture of CNN and Transformer, for the building type classification from Very-High-Resolution (VHR) remote sensing images. By merging two techniques, we can effectively enhance the learning of intricate spatial features and detailed characteristics of buildings within complex scenarios. Moreover, we created a building type dataset, i.e., an Urban Building Type dataset (UBTD) to validate the effectiveness of our method, addressing the scarcity of datasets in current research. Extensive experiments conducted on the UBTD datasets demonstrate that HFINet outperforms popular CNN, Transformer methods, and dual encoder methods in building type classification across various regions, achieving the highest MIOU values of 72.20%, F1 values of 80.83% and ACC of 93.07%. Our approach, HFINet, excels in very-high-resolution remote sensing building type classification, offering improved accuracy and robustness, even in complex environments.

关键词： Accuracy Smart cities Architecture Buildings Semantics Estimation Carbon dioxide Transformers Robustness Remote sensing

来源：评论

学校读者我要写书评

暂无评论

Urban Land Use Change Detection Using Siamese Graph Neural Networks from Bi-Temporal Very-High-Resolution Satellite Images

Urban Land Use Change Detection Using Siamese Graph Neural N...

引用

Artificial Intelligence Technology (ACAIT), Asian Conference on

作者： Kangkai Lou Mengmeng Li Shaofeng Zhang Key Lab of Spatial Data Mining & Information Sharing of Ministry of Education Academy of Digital China (Fujian) Fuzhou University Fuzhou China

ISBN: (数字)9798331517090

ISBN: (纸本)9798331517106

Accurate land use change information is crucial for sustainable urban development, environmental conservation, public health, and socio-economic well-being. This paper explores the application of bi-temporal very-high-resolution (VHR) remote sensing images for detecting land use changes. In this paper, we introduced a novel network specifically designed for this task, termed the Siamese Graph Neural Network (SGNN-Net), which represents scene images into graph-structured data and uses graph neural networks (GNNs) to extract high-level information among land cover objects to assist in the detection of land use changes. We conducted experiments on a public HY dataset and a newly created FZ dataset. Our findings indicate that the proposed method performs effectively across both datasets, demonstrating high agreement between detected land use changes and ground-truth change maps. This suggests that the method holds significant promise for detecting land use changes in urban areas using bi-temporal VHR images, even in regions with a notable imbalance between changed and unchanged samples.

关键词： Urban areas Land surface Feature extraction Graph neural networks Satellite images data mining Socioeconomics Public healthcare Environmental management Remote sensing

来源：评论

学校读者我要写书评

暂无评论

Prediction of river water quality considering spatiotemporal correlation and meteorological factors 9

Prediction of river water quality considering spatiotemporal...

引用

9th International Conference on Agro-Geoinformatics, Agro-Geoinformatics 2021

作者： Lu, Yi-Min Zhang, Hong Shao, Wei Fuzhou University Fuzhou Key Lab of Spatial Data Mining and Information Sharing of Ministry of Education Fujian350108 China

ISBN: (纸本)9781728165615

With the rapid development of economy, the problem of river pollution is becoming more and more serious. It is very important and challenging to build a high-precision water quality prediction model for the comprehensive management of water environment and prevention of water pollution. At present, some achievements have been made in data-driven river water quality prediction, mainly including grey model, time series model, support vector machine model and neural network model. However, these existing prediction methods are limited to single station prediction. The influence of the upstream water quality on the downstream water quality is ignored and the upstream stations at different locations will have different effects on the downstream water quality. In order to solve the problem that the traditional model of water quality prediction does not consider the upstream influence and difficulties in long-term prediction, a new water quality prediction model based on spatiotemporal attention mechanism and long-short-term memory neural network (TS-Attention-LSTM) was proposed, which considers the spatiotemporal correlation and meteorological factors. Firstly, the spatial attention module was embedded in the encoder to extract the significant spatial correlation between upstream and downstream. The interaction among the water quality indicators was also extracted. Then, the temporal attention module was embedded in the decoder to extract the important time series features. Moreover, the meteorological factors and spatial characteristics were fused in the decoder. Finally, the multi-step prediction of water quality was carried out by using LSTM model. In this paper, the research area located in Jinjiang River Watershed in Fujian Province, China. The nonpoint source pollution in this river basin mainly interrelated with livestock discharge, agricultural field runoff and rural domestic sewage. The results show that the TS-Attention-LSTM model can effectively capture the spa

关键词： Water quality

来源：评论

学校读者我要写书评

暂无评论

Integration of CNN and Transformer for Regularized Extraction of Agricultural Parcels from High-Resolution Remote Sensing Images

Integration of CNN and Transformer for Regularized Extractio...

引用

Artificial Intelligence Technology (ACAIT), Asian Conference on

作者： Chengwen Lu Mengmeng Li Xiaolong Xiu Jiang Long Congcong Zhang Key Lab of Spatial Data Mining & Information Sharing of Ministry of Education Academy of Digital China (Fujian) Fuzhou University FuZhou China Geological Remote Sensing Center Fujian Geological Survey and Mapping Institute FuZhou China

ISBN: (数字)9798331517090

ISBN: (纸本)9798331517106

Agricultural parcels are the fundamental spatial units of precision agricultural practices. However, in southern regions, the parcels tend to be small, structurally complex, and fragmented, posing challenges for precise extraction using existing convolutional neural network (CNNs) based on deep learning. To address these issues, we introduced a novel neural network model, i.e., CLCFormer, which combines CNN and Swin-Transformer. This model leverages the strengths of CNN in capturing fine-grained spatial details and the ability of Transformer in extracting global semantic features. We introduced three distinct modules to refine the learning of spatial features, enhance the accuracy of parcel extraction in mountainous and hilly regions, and mitigate the boundary blurriness caused by occluded objects. Finally, we employed a post-processing method to achieve regularized parcel extraction. Our study was tested on GF-2 high-resolution remote sensing datasets from Nanping and Fuzhou, Fujian Province, China. Our proposed method achieved a prediction precision of 86.68% and a recall rate of 85.3% on the Nanping test set, and maintaining a transfer precision of 85.67% on the Fuzhou dataset. Compared to other advanced parcel extraction models, our method demonstrates the highest performance. Moreover, the extracted parcels have the highest consistency with the actual distribution patterns of agricultural parcels. We provided a practical approach for precise extraction of agricultural parcels information at the parcel scale.

关键词： Deep learning Semantics Neural networks Vegetation mapping Feature extraction Transformers Vectors Robustness Convolutional neural networks Remote sensing

来源：评论

学校读者我要写书评

暂无评论

Differential Lightweight Processing in Digital Twin Construction of Substations

Differential Lightweight Processing in Digital Twin Construc...

引用

Energy Engineering and Power Systems (EEPS), International Conference on

作者： Meng Zhang Ding Lin Jing Ding Tao Fang Key Lab of Spatial Data Mining and Information Sharing of Ministry of Education Fuzhou University Fuzhou Fujian China State Grid ZHEJIANG Electric Power co. LTD State Grid Hangzhou Power Supply Company Hangzhou Zhejiang China

ISBN: (数字)9798350375138

ISBN: (纸本)9798350375145

This study aims to address the challenge of lightweighting point cloud data at the part level in substation equipment, aligning with the requirements of digital twin applications for grid substations and considering the hierarchical structure of equipment. Utilizing the importance of equipment parts, we construct a priority queue and integrate the Draco algorithm to automatically lightweight transformer equipment point cloud into semantic models with multiple levels of detail at the part level. This approach effectively resolves the lightweighting issue of point cloud data for substation equipment at the part level, achieving a differential treatment where key parts maintain high original data retention rates while minor parts occupy fewer resources. In comparison to applying the global Draco algorithm to equipment, our method offers the advantage of flexibly allocating storage and transmission resources to user-focused key targets while keeping the overall compression rate unchanged.

关键词： Point cloud compression Substations Semantics Energy resolution Rate distortion theory Transformers Distortion data models Digital twins Smart grids

来源：评论

学校读者我要写书评

暂无评论

Evaluating the point cloud of individual trees generated from images based on Neural Radiance fields (NeRF) method

arXiv

引用

arXiv 2023年

作者： Huang, Hongyu Tian, Guoji Chen, Chongcheng National Engineering Research Center of Geospatial Information Technology Fuzhou University Fuzhou350108 China Key Lab of Spatial Data Mining and Information Sharing of Ministry of Education Fuzhou University Fuzhou350108 China Fuzhou350108 China

Three-dimensional (3D) reconstruction of trees has always been a key task in precision forestry management and research. Due to the complex branch morphological structure of trees themselves and the occlusions from tree stems, branches and foliage, it is difficult to recreate a complete three-dimensional tree model from a two-dimensional image by conventional photogrammetric methods. In this study, based on tree images collected by various cameras in different ways, the Neural Radiance Fields (NeRF) method was used for individual tree reconstruction and the exported point cloud models are compared with point cloud derived from photogrammetric reconstruction and laser scanning methods. The results show that the NeRF method performs well in individual tree 3D reconstruction, as it has higher successful reconstruction rate, better reconstruction in the canopy area, it requires less amount of images as input. Compared with photogrammetric reconstruction method, NeRF has significant advantages in reconstruction efficiency and is adaptable to complex scenes, but the generated point cloud tends to be noisy and low resolution. The accuracy of tree structural parameters (tree height and diameter at breast height) extracted from the photogrammetric point cloud is still higher than those of derived from the NeRF point cloud. The results of this study illustrate the great potential of NeRF method for individual tree reconstruction, and it provides new ideas and research directions for 3D reconstruction and visualization of complex forest scenes. Copyright © 2023, The Authors. All rights reserved.

关键词： Photogrammetry

来源：评论

学校读者我要写书评

暂无评论

Developing a new red band- SEVI-blue band (RSB) enhancement method for recognition the extra-high-voltage transmission line corridor in green mountains

引用

International Journal of Digital Earth 2023年第1期16卷 806-824页

作者： Hong Jiang Yong Zhang Jinglan Lin Xiaogan Zheng Hui Yue Yunzhi Chen Key Laboratory of Spatial Data Mining&Information Sharing of Ministry of Education National&Local Joint Engineering Research Center of Satellite Geospatial Information TechnologyAcademy of Digital China(Fujian)Fuzhou UniversityFuzhouPeople’s Republic of China Fujian Soil and Water Conservation Experimental Station FuzhouPeople’s Republic of China Lab of Live Working Tech for Power Transmission Fuzhou Power Supply Company of State Grid Fujian Electric Power Co.Ltd.FuzhouPeople’s Republic of China Fujian(Changting)Soil and Water Conservation Research Center Soil and Water Conservation Center of Changting CountyLongyanPeople’s Republic of China

Monitoring the extra-high-voltage transmission line corridor(EHVTLC)in mountains is critical for safe smart-grid ***,the transmission lines are so narrow that they are difficult to recognize using multispectral satellite images with a spatial resolution of 10 *** this study,we developed a new method using the red band–shadow-eliminated vegetation index(SEVI)–blue band(RSB)composite image to enhance the EHVTLC in green mountains(named RSB-enhancement method).Using this method,the EHVTLC becomes evident in the false-color synthesis of the RSB composite of the Sentinel-2 ***,we recognized and extracted approximately 342.45 km of the EHVTLC in a mountainous region of Fuzhou City,China,including a 46.73 km three-parallel-lane segment of 1000 kV and a 295.72 km two-parallel-lane segment of 500 *** analysis shows that the SEVI mean difference between the EHVTLC and the buffer zone reaches approximately 10%,and three landslides and 2.66 km^(2) soil erosion reside in the buffer zone which area is approximately 73.67 km^(2).Finally,the RSB-enhancement method can be used in other satellite images with spatial resolutions of greater than 10 m for enhancement and recognition the transmission line corridors in green mountains.

关键词： Enhancement method shadow-eliminated vegetation index(SEVI) transmission line lane(TLL) green mountains soil erosion

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：