Filters








7,427 Hits in 4.4 sec

Scale-Robust Deep-Supervision Network for Mapping Building Footprints from High-Resolution Remote Sensing Images

Haonan Guo, Xin Su, Shengkun Tang, Bo Du, Liangpei Zhang
2021 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
To tackle these limitations, we propose a novel deep-supervision convolutional neural network (denoted as DS-Net) for extracting building footprints from high-resolution remote sensing images.  ...  Building footprint information is one of the key factors for sustainable urban planning and environmental monitoring.  ...  To tackle these problems, in this article, we designed a novel deep-supervision fully convolutional network (denoted as DS-Net) for building footprint extraction.  ... 
doi:10.1109/jstars.2021.3109237 fatcat:4qfx36ge3nev7p2fq2cjrtxlp4

PP-LinkNet: Improving Semantic Segmentation of High Resolution Satellite Imagery with Multi-stage Training

An Tran, Ali Zonoozi, Jagannadan Varadarajan, Hannes Kruppa
2020 Proceedings of the 2nd Workshop on Structuring and Understanding of Multimedia heritAge Contents  
Road network and building footprint extraction is essential for many applications such as updating maps, traffic regulations, city planning, ride-hailing, disaster response etc.  ...  We further propose Pyramid Pooling-LinkNet (PP-LinkNet), an improved deep neural network for segmentation that uses focal loss, poly learning rate, and context module.  ...  First, we develop a novel method to generate pseudo road network and building footprint ground truth masks from OSM data without human annotation labor.  ... 
doi:10.1145/3423323.3423407 dblp:conf/mm/TranZVK20 fatcat:4vo372ozpnecjchhmwsz7oacge

PiCoCo: Pixelwise Contrast and Consistency Learning for Semi-Supervised Building Footprint Segmentation

Jian Kang, Zhirui Wang, Ruoxin Zhu, Xian Sun, Ruben Fernandez-Beltran, Antonio J Plaza
2021 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
In this paper, we propose a novel semisupervised learning method for building footprint segmentation, which can effectively predict building footprints based on the network trained with few annotations  ...  Convolutional neural networks (CNNs) have been recently used as a workhorse for effectively generating building footprints.  ... 
doi:10.1109/jstars.2021.3119286 fatcat:7xfdzrevnjeehbqpw5z2k45xq4

Fine Building Segmentation in High-Resolution SAR Images via Selective Pyramid Dilated Network

Hao Jing, Xian Sun, Zhirui Wang, Kaiqiang Chen, Wenhui Diao, Kun Fu
2021 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
Firstly, we propose a novel encoderdecoder structure for the fine building feature reconstruction.  ...  Recently, the deep convolution neural network brings excellent improvements in SAR segmentation.  ...  The authors are very grateful for the helpful suggestions of anonymous reviewers.  ... 
doi:10.1109/jstars.2021.3076085 fatcat:3gzb3c2jbfaxnfnzayoteofmcq

Topological Map Extraction from Overhead Images [article]

Zuoyue Li, Jan Dirk Wegner, Aurélien Lucchi
2019 arXiv   pre-print
PolyMapper directly extracts the topological map of a city from overhead images as collections of building footprints and road networks.  ...  In order to unify the shape representation for different types of objects, we also propose a novel sequentialization method that reformulates a graph structure as closed polygons.  ...  for only one of the tasks, road network prediction [32, 4] or building footprint extraction [38] .  ... 
arXiv:1812.01497v3 fatcat:bn4elkzm3vgr3i5vrofgsb57fq

MAP-Net: Multi Attending Path Neural Network for Building Footprint Extraction from Remote Sensed Imagery [article]

Qing Zhu, Cheng Liao, Han Hu, Xiaoming Mei, Haifeng Li
2019 arXiv   pre-print
This paper proposes a novel multi attending path neural network (MAP-Net) for accurately extracting multiscale building footprints and precise boundaries.  ...  Then, an attention module adaptively squeezes channel-wise features from each path for optimization, and a pyramid spatial pooling module captures global dependency for refining discontinuous building  ...  , in this research, we proposed a novel localization-preserved multipath feature extraction network with a channel and spatial enhancement module for building footprint extraction.  ... 
arXiv:1910.12060v1 fatcat:zdzgutvrjjbyhdnqsrplipfojm

ADVANCED APPROACH FOR AUTOMATIC RECONSTRUCTION OF 3D BUILDINGS FROM AERIAL IMAGES

D. Yu, S. Wei, J. Liu, S. Ji
2020 The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
In this work, a novel automatic 3D building reconstruction approach is proposed to extract accurate LoD1 building models from multi-view aerial images.  ...  The second step is to produce structured 2D building footprints using combined deep learning and regularization.  ...  Due to the significant advances in deep learning, the convolutional neural network (CNN) based semantic segmentation methods have increasingly been used for building footprint extraction from remote sensing  ... 
doi:10.5194/isprs-archives-xliii-b2-2020-541-2020 fatcat:xsmih6h7r5cylp3y4wncyzhkie

A Histogram Thresholding Improvement to Mask R-CNN for Scalable Segmentation of New and Old Rural Buildings [article]

Ying Li, Weipan Xu, Haohui Chen, Junhao Jiang, Xun Li
2021 arXiv   pre-print
In recent years, deep neural networks have achieved remarkable building segmentation results in high-resolution remote sensing images.  ...  However, the scarce training data and the varying geographical environments have posed challenges for scalable building segmentation.  ...  As discussed before, we propose a novel segmentation framework that can utilize the histogram thresholding and deep learning's image segmentation capability to extract the new and old rural buildings.  ... 
arXiv:2102.04838v1 fatcat:suh3eutwvbgkxiqqv5z7fbwawu

CG-Net: Conditional GIS-Aware Network for Individual Building Segmentation in VHR SAR Images

Yao Sun, Yuansheng Hua, Lichao Mou, Xiao Xiang Zhu
2021 IEEE Transactions on Geoscience and Remote Sensing  
To achieve this, we introduce building footprints from geographic information system (GIS) data as a complementary information and propose a novel conditional GIS-aware network (CG-Net).  ...  We further compare two representations of building footprints, namely, complete building footprints and sensor-visible footprint segments, for our task, and conclude that the use of the former leads to  ...  Hirschmüller of DLR-RM for providing the optical DEM.  ... 
doi:10.1109/tgrs.2020.3043089 fatcat:yf4l6fjzbfa2bk5jz63ic6ngwu

A Simple But Effective Approach of Building Footprint Extraction in Topographic Mapping Acceleration

Danang Budi Susetyo, Aldino Rizaldy, Mochamad Irwan Hariyono, Nugroho Purwono, Fahrul Hidayat, Rizka Windiastuti, Tia Rizka N. Rachma, Prayudha Hartanto
2021 Indonesian Journal on Geoscience  
Thus, this research was conducted to find the effective way to extract building footprint for mapping acceleration from LiDAR data.  ...  Segmentation using segment growing was used to separate each building, so boundary detection could be conducted for each segment to create boundary of each building.  ...  A recent technology is based on deep learning algorithms, for example using neural networks and Markov Random Fields (MRF) (Davydova et al., 2016) , end-to-end trainable gated residual refinement network  ... 
doi:10.17014/ijog.8.3.329-343 fatcat:453d5yzn3fasrikjbhkmerj5bi

Optimization of OpenStreetMap Building Footprints Based on Semantic Information of Oblique UAV Images

Xiangyu Zhuo, Friedrich Fraundorfer, Franz Kurz, Peter Reinartz
2018 Remote Sensing  
In parallel, a deep neural network for pixel-wise semantic image segmentation is trained in order to extract the building boundaries as contour evidence.  ...  The contour evidence is extracted from pixel-wise semantic segmentation via deep convolution neural networks.  ...  Acknowledgments: This research was funded by the German Academic Exchange Service (DAAD:DLR/DAAD Research Fellowship Nr. 50019750) for Xiangyu Zhuo.  ... 
doi:10.3390/rs10040624 fatcat:6chqxgzo6ng2pnd6uorymbvsya

CG-Net: Conditional GIS-aware Network for Individual Building Segmentation in VHR SAR Images [article]

Yao Sun, Yuansheng Hua, Lichao Mou, Xiao Xiang Zhu
2020 arXiv   pre-print
To achieve this, we introduce building footprints from GIS data as complementary information and propose a novel conditional GIS-aware network (CG-Net).  ...  We further compare two representations of building footprints, namely complete building footprints and sensor-visible footprint segments, for our task, and conclude that the use of the former leads to  ...  Hirschmüller of DLR-RM for providing the optical DEM.  ... 
arXiv:2011.08362v1 fatcat:oc6cq26bdjcdhapasyu7lefxoi

Effective Building Extraction by Learning to Detect and Correct Erroneous Labels in Segmentation Mask

Praveer Singh, Nikos Komodakis
2018 IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium  
We evaluate our methodology on the benchmark Inria Aerial Image Labeling dataset, which is a large scale high resolution dataset for building footprint segmentation.  ...  In this paper, we address this problem and propose a novel solution by modeling the joint distribution of input-output variable which in turn enforces some structure in the initial segmentation mask.  ...  Network Architecture We initially train a Fully Convolutional Network [11] (FCN) adapted to Resnet-50 [12] architecture to generate our initial segmentation mask Y .  ... 
doi:10.1109/igarss.2018.8517854 dblp:conf/igarss/SinghK18 fatcat:eo7gopjldjbxlhbav5ilr4onky

An Evacuation Route Model for Disaster Affected Areas

Vinaysheel K. Wagh, Pramod Pathak, Paul Stynes, Luis G. Nardin
2020 Irish Conference on Artificial Intelligence and Cognitive Science  
This paper proposes a novel model for classifying damaged buildings and supporting people's evacuation from natural disaster affected areas using satellite images.  ...  The model integrates image segmentation and classification with a shortest path algorithm. First, buildings are detected from pre-disaster satellite images using the proposed Segmentation model.  ...  Model process Fig. 4 : 4 Fig. 4: Route Detection Model output Fig. 5 : 5 Fig. 5: Evacuation route adaptability Table 1 : 1 Parameter settings for the Building Footprint Extraction and Segmentation  ... 
dblp:conf/aics/WaghPSN20 fatcat:5o2mfktbqvbclitsmku26awrpm

RescueNet: Joint Building Segmentation and Damage Assessment from Satellite Imagery [article]

Rohit Gupta, Mubarak Shah
2020 arXiv   pre-print
In order to to model the composite nature of this problem, we propose a novel localization aware loss function, which consists of a Binary Cross Entropy loss for building segmentation, and a foreground  ...  We propose RescueNet, a unified model that can simultaneously segment buildings and assess the damage levels to individual buildings and can be trained end-toend.  ...  The dataset uses polygons to reprsent building segments and a novel 4 point damage scale.  ... 
arXiv:2004.07312v1 fatcat:tkfp2mfaifb4jfv5ufnv7q4fqe
« Previous Showing results 1 — 15 out of 7,427 results