Centre de documentation
scientifique

Feature-selection high-resolution network with hypersphere embedding for semantic segmentation of VHR remote sensing images / Hanwen Xu in IEEE Transactions on geoscience and remote sensing, vol 60 n° 6 (June 2022)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 60 n° 6 (June 2022) . - n° 4411915 Titre : Feature-selection high-resolution network with hypersphere embedding for semantic segmentation of VHR remote sensing images Type de document : Article/Communication Auteurs : Hanwen Xu, Auteur ; Xinming Tang, Auteur ; Bo Ai, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : n° 4411915 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] apprentissage profond [Termes IGN] architecture de réseau [Termes IGN] attention (apprentissage automatique) [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] entropie [Termes IGN] extraction de traits caractéristiques [Termes IGN] image à très haute résolution [Termes IGN] segmentation multi-échelle [Termes IGN] segmentation sémantique Résumé : (auteur) Very-high-resolution (VHR) remote sensing images contain various multiscale objects, such as large-scale buildings and small-scale cars. However, these multiscale objects cannot be considered simultaneously in the widely used backbones with a large downsampling factor (e.g., VGG-like and ResNet-like), resulting in the appearance of various context aggregation approaches, such as fusing low-level features and attention-based modules. To alleviate this problem caused by backbones with a large downsampling factor, we propose a feature-selection high-resolution network (FSHRNet) based on an observation: if the features maintain high resolution throughout the network, a high precision segmentation result can be obtained by only using a 1× 1 convolution layer with no need for complex context aggregation modules. Specifically, the backbone of FSHRNet is a multibranch structure similar to HRNet where the high-resolution branch is the principal line. Then, a lightweight dynamic weight module, named the feature-selection convolution (FSConv) layer, is presented to fuse multiresolution features, allowing adaptively feature selection based on the characteristic of objects. Finally, a specially designed 1× 1 convolution layer derived from hypersphere embedding is used to produce the segmentation result. Experiments with other widely used methods show that the proposed FSHRNet obtains competitive performance on the ISPRS Vaihingen dataset, the ISPRS Potsdam dataset, and the iSAID dataset. Numéro de notice : A2022-559 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2022.3183144 En ligne : https://doi.org/10.1109/TGRS.2022.3183144 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=101184 [article]

Integration of ZY3-02 satellite laser altimetry data and stereo images for high-accuracy mapping / Guoyuan Li in Photogrammetric Engineering & Remote Sensing, PERS, vol 84 n° 9 (September 2018)  

Public [article]  inPhotogrammetric Engineering & Remote Sensing, PERS > vol 84 n° 9 (September 2018) . - pp 569 - 578 Titre : Integration of ZY3-02 satellite laser altimetry data and stereo images for high-accuracy mapping Type de document : Article/Communication Auteurs : Guoyuan Li, Auteur ; Xinming Tang, Auteur ; Xiaoming Gao, Auteur ; et al., Auteur Année de publication : 2018 Article en page(s) : pp 569 - 578 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image [Termes IGN] Chine [Termes IGN] données altimétriques [Termes IGN] données ICEsat [Termes IGN] image ZiYuan-3 [Termes IGN] modèle par fonctions rationnelles [Termes IGN] ZiYuan-3 Résumé : (Auteur) Integration of satellite laser altimetry data and stereo images without ground control points (GCPs) is an attractive method for global mapping. In this paper, we propose a new strategy of integrating Ziyuan3-02 (ZY3-02) satellite stereo images and laser altimetry data using a rigorous sensor model (RSM) with laser ranging constraint under the synchronized and rational function model (RFM) with laser elevation constraint under the non-synchronized capture for high-accuracy mapping without GCPs. Four experimental regions in China are selected to validate the method. The results show that the ZY3-02 satellite laser altimetry data can be used to improve the elevation accuracy of stereo images to better than 3.0 m without GCPs. All of the conclusions are valuable for the development of China's next generation of surveying and mapping satellites. Numéro de notice : A2018-362 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.14358/PERS.84.9.569 Date de publication en ligne : 01/09/2018 En ligne : https://doi.org/10.14358/PERS.84.9.569 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=90673 [article]

Exemplaires(1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
105-2018091	RAB	Revue	Centre de documentation	En réserve L003	Disponible

ZY-3 block adjustment supported by GLAS laser altimetry data / Guoyuan Li in Photogrammetric record, vol 31 n° 153 (March - May 2016)  

Public [article]  inPhotogrammetric record > vol 31 n° 153 (March - May 2016) . - pp 88 - 107 Titre : ZY-3 block adjustment supported by GLAS laser altimetry data Type de document : Article/Communication Auteurs : Guoyuan Li, Auteur ; Xinming Tang, Auteur ; Xiaoming Gao, Auteur ; et al., Auteur Année de publication : 2016 Article en page(s) : pp 88 - 107 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image mixte [Termes IGN] altimétrie satellitaire par laser [Termes IGN] compensation par bloc [Termes IGN] données altimétriques [Termes IGN] données ICEsat [Termes IGN] image ZiYuan-3 Résumé : (Auteur) ZY-3 is the first Chinese civilian high-resolution stereo-mapping satellite, designed for 1:50 000 scale mapping. Although ZY-3 has a very high accuracy for direct image geolocation without ground control points (GCPs), the use of some GCPs is still indispensable for stereo mapping. A novel method is proposed whereby spaceborne Geoscience Laser Altimeter System (GLAS) laser altimetry datasets are viewed as generalised height control points to support a ZY-3 block adjustment. This method separates planimetric and height control, integrating the latter with GLAS data such that, without other GCPs, the vertical accuracy of ZY-3 images can be improved significantly. Both flat and mountainous terrain were selected to evaluate the proposed method. The results show that by adding two GLAS height control points for a single image, or spaced every 50 km in the along-track (orbital) direction for block images, the vertical accuracy can satisfy the requirements of 1:50 000 scale mapping, thus reducing the need for field surveying. Additionally, the method will be valuable for integrating spaceborne laser altimetry data and optical images for global mapping without any GCPs. Numéro de notice : A2016-163 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.1111/phor.12138 Date de publication en ligne : 01/03/2016 En ligne : https://doi.org/10.1111/phor.12138 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=80464 [article]

Exemplaires(1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
106-2016011	RAB	Revue	Centre de documentation	En réserve L003	Disponible

Geometric accuracy analysis model of the ZiYuan-3 satellite without GCPs / Xinming Tang in Photogrammetric Engineering & Remote Sensing, PERS, vol 81 n° 12 (December 2015)  

Public [article]  inPhotogrammetric Engineering & Remote Sensing, PERS > vol 81 n° 12 (December 2015) . - pp 927 - 934 Titre : Geometric accuracy analysis model of the ZiYuan-3 satellite without GCPs Type de document : Article/Communication Auteurs : Xinming Tang, Auteur ; Ping Zhou, Auteur ; Guo Zhang, Auteur ; et al., Auteur Année de publication : 2015 Article en page(s) : pp 927 - 934 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Acquisition d'image(s) et de donnée(s) [Termes IGN] estimation des paramètres [Termes IGN] évaluation des données [Termes IGN] évaluation des paramètres [Termes IGN] image ZiYuan-3 [Termes IGN] modèle géométrique de prise de vue [Termes IGN] précision géométrique (imagerie) [Termes IGN] propagation d'erreur [Termes IGN] source d'erreur Résumé : (auteur) The ZiYuan-3 satellite (ZY-3) was China’s first civilian stereo mapping satellite designed to meet the 1:50 000 scale mapping requirements. We analyzed main error sources influencing the geometric accuracy of the ZY-3 images, and the error propagation rules of these error sources in image production. Accuracy estimation models of ZY-3 images without ground control points (GCPs) were deduced and the theoretical geometric accuracy of ZY-3 images was obtained. Without GCPs, 556 ZY-3 panchromatic nadir images covering 3,500,000 km2 of the Chinese mid-west region and 12 ZY-3 stereo image pairs covering 14,000 km2 around Taiyuan of China were used for planar and vertical accuracy verification, respectively. The experimental results confirmed the correctness of the accuracy estimation model for ZY-3 images. The accuracy results obtained through the model and experiment showed that without GCPs, the geometric accuracy of the ZY-3 images satisfied the Chinese stereo mapping requirements for 1:50 000 scale topographic maps Numéro de notice : A2015-992 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.14358/PERS.81.12.927 En ligne : https://doi.org/10.14358/PERS.81.12.927 Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=80273 [article]