Centre dedocumentation
scientifique

A lightweight network with attention decoder for real-time semantic segmentation / Kang Wang in The Visual Computer, vol 38 n° 7 (July 2022)  

Public [article]  inThe Visual Computer > vol 38 n° 7 (July 2022) . - pp 2329 - 2339 Titre : A lightweight network with attention decoder for real-time semantic segmentation Type de document : Article/Communication Auteurs : Kang Wang, Auteur ; Jinfu Yang, Auteur ; Shuai Yuan, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : pp 2329 - 2339 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] attention (apprentissage automatique) [Termes IGN] jeu de données [Termes IGN] précision [Termes IGN] segmentation sémantique [Termes IGN] temps réel [Termes IGN] vitesse de traitement Résumé : (auteur) As an important task in scene understanding, semantic segmentation requires a large amount of computation to achieve high performance. In recent years, with the rise of autonomous systems, it is crucial to make a trade-off in terms of accuracy and speed. In this paper, we propose a novel asymmetric encoder–decoder network structure to address this problem. In the encoder, we design a Separable Asymmetric Module, which combines depth-wise separable asymmetric convolution with dilated convolution to greatly reduce computation cost while maintaining accuracy. On the other hand, an attention mechanism is also used in the decoder to further improve segmentation performance. Experimental results on CityScapes and CamVid datasets show that the proposed method can achieve a better balance between segmentation precision and speed compared with state-of-the-art semantic segmentation methods. Specifically, our model obtains mean IoU of 72.5% and 66.3% on CityScapes and CamVid test dataset, respectively, with less than 1M parameters. Numéro de notice : A2022-508 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.1007/s00371-021-02115-4 Date de publication en ligne : 07/05/2021 En ligne : https://doi.org/10.1007/s00371-021-02115-4 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=101041 [article]

Semantic feature-constrained multitask siamese network for building change detection in high-spatial-resolution remote sensing imagery / Qian Shen in ISPRS Journal of photogrammetry and remote sensing, vol 189 (July 2022)  

Public [article]  inISPRS Journal of photogrammetry and remote sensing > vol 189 (July 2022) . - pp 78 - 94 Titre : Semantic feature-constrained multitask siamese network for building change detection in high-spatial-resolution remote sensing imagery Type de document : Article/Communication Auteurs : Qian Shen, Auteur ; Jiru Huang, Auteur ; Min Wang, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : pp 78 - 94 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] détection de changement [Termes IGN] détection du bâti [Termes IGN] données qualitatives [Termes IGN] estimation quantitative [Termes IGN] fusion d'images [Termes IGN] image à haute résolution [Termes IGN] image multibande [Termes IGN] jeu de données [Termes IGN] réseau neuronal siamois Résumé : (auteur) In the field of remote sensing applications, semantic change detection (SCD) simultaneously identifies changed areas and their change types by jointly conducting bitemporal image classification and change detection. It facilitates change reasoning and provides more application value than binary change detection (BCD), which offers only a binary map of the changed/unchanged areas. In this study, we propose a multitask Siamese network, named the semantic feature-constrained change detection (SFCCD) network, for building change detection in bitemporal high-spatial-resolution (HSR) images. SFCCD conducts feature extraction, semantic segmentation and change detection simultaneously, where change detection and semantic segmentation are the main and auxiliary tasks, respectively. For the segmentation task, ResNet50 is used to conduct image feature extraction, and the extracted semantic features are provided to execute the change detection task via a series of jump connections. For the change detection task, a global channel attention (GCA) module and a multiscale feature fusion (MSFF) module are designed, where high-level features offer training guidance to the low-level feature maps, and multiscale features are fused with multiple convolutions that possess different receptive fields. In bitemporal HSR images with different view angles, high-rise buildings have different directional height displacements, which generally cause serious false alarms for common change detection methods. However, known public building change detection datasets often lack buildings with height displacement. We thus create the Nanjing Dataset (NJDS) and design the aforementioned network structures and modules to target this issue. Experiments for method validation and comparison are conducted on the NJDS and two additional public datasets, i.e., the WHU Building Dataset (WBDS) and Google Dataset (GDS). Ablation experiments on the NJDS show that the joint utilization of the GCA and MSFF modules performs better than several classic modules, including atrous spatial pyramid pooling (ASPP), efficient spatial pyramid (ESP), channel attention block (CAB) and global attention upsampling (GAU) modules, in dealing with building height displacement. Furthermore, SFCCD achieves higher accuracy in terms of the OA, recall, F1-score and mIoU measures than several state-of-the-art change detection methods, including deeply supervised image fusion network (DSIFN), the dual-task constrained deep Siamese convolutional network (DTCDSCN), and multitask U-Net (MTU-Net). Numéro de notice : A2022-412 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.isprsjprs.2022.05.001 Date de publication en ligne : 12/05/2022 En ligne : https://doi.org/10.1016/j.isprsjprs.2022.05.001 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=100762 [article]

Réservation

Réserver ce document

Exemplaires (1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
081-2022071	SL	Revue	Centre de documentation	Revues en salle	Disponible

A cost-effective algorithm for calibrating multiscale geographically weighted regression models / Bo Wu in International journal of geographical information science IJGIS, vol 36 n° 5 (May 2022)  

Public [article]  inInternational journal of geographical information science IJGIS > vol 36 n° 5 (May 2022) . - pp 898 - 917 Titre : A cost-effective algorithm for calibrating multiscale geographically weighted regression models Type de document : Article/Communication Auteurs : Bo Wu, Auteur ; Jinbiao Yan, Auteur ; Hui Lin, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : pp 898 - 917 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Analyse spatiale [Termes IGN] analyse multiéchelle [Termes IGN] grande échelle [Termes IGN] hétérogénéité spatiale [Termes IGN] jeu de données [Termes IGN] modélisation spatiale [Termes IGN] régression géographiquement pondérée Résumé : (auteur) The multiscale geographically weighted regression (MGWR) model is a useful extension of the geographically weighted regression (GWR) model. MGWR, however, is a kind of Nadaraya–Watson kernel smoother, which usually leads to inaccurate estimates for the regression function and suffers from the boundary effect. Moreover, the widely used calibration technique for the MGWR with a back-fitting estimator (MGWR-BF) is computationally demanding, preventing it from being applied to large-scale data. To overcome these problems, we proposed a local linear-fitting-based MGWR (MGWR-LL) by introducing a local spatially varying coefficient model in which coefficients of different variables could be characterised as linear functions of spatial coordinates with different degrees of smoothness. Then the model was calibrated with a two-step least-squared estimated algorithm. Both simulated and actual data were implemented to validate the performance of the proposed method. The results consistently showed that the MGWR-LL automatically corrected for the boundary effect and improved the accuracy in most cases, not only in the goodness-of-fit measure but also in reducing the bias of the coefficient estimates. Moreover, the MGWR-LL significantly outperformed the MGWR-BF in computational cost, especially for larger-scale data. These results demonstrated that the proposed method can be a useful tool for the MGWR calibration. Numéro de notice : A2022-342 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1080/13658816.2021.1999457 Date de publication en ligne : 29/11/2021 En ligne : https://doi.org/10.1080/13658816.2021.1999457 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=100516 [article]

Deep mass redistribution prior to the 2010 Mw 8.8 Maule (Chile) Earthquake revealed by GRACE satellite gravity / Marie Bouih in Earth and planetary science letters, vol 584 (15 April 2022)  

Public [article]  inEarth and planetary science letters > vol 584 (15 April 2022) . - n° 117465 Titre : Deep mass redistribution prior to the 2010 Mw 8.8 Maule (Chile) Earthquake revealed by GRACE satellite gravity Type de document : Article/Communication Auteurs : Marie Bouih , Auteur ; Isabelle Panet , Auteur ; Dominique Remy, Auteur ; Laurent Longuevergne, Auteur ; Sylvain Bonvalot, Auteur Année de publication : 2022 Projets : Université de Paris / Clerici, Christine Article en page(s) : n° 117465 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Géodésie physique [Termes IGN] champ de pesanteur terrestre [Termes IGN] Chili [Termes IGN] déformation de la croute terrestre [Termes IGN] données GRACE [Termes IGN] gradient de gravitation [Termes IGN] jeu de données [Termes IGN] levé gravimétrique [Termes IGN] prévention des risques [Termes IGN] risque naturel [Termes IGN] séisme [Termes IGN] signal [Termes IGN] subduction [Termes IGN] tectonique des plaques Résumé : (auteur) Subduction zones megathrust faults constitute a considerable hazard as they produce most of the world's largest earthquakes. However, the role in megathrust earthquake generation exerted by deeper subduction processes remains poorly understood. Here, we analyze the 2003 – 2014 space-time variations of the Earth's gravity gradients derived from three datasets of GRACE geoid models over a large region surrounding the rupture zone of the Mw 8.8 Maule earthquake. In all these datasets, our analysis reveals a large-amplitude gravity gradient signal, progressively increasing in the three months before the earthquake, North of the epicentral area. We show that such signals are equivalent to a water storage decrease over 2 months and cannot be explained by hydrological sources nor artefacts, but rather find origin from mass redistributions within the solid Earth on the continental side of the subduction zone. These gravity gradient variations could be explained by an extensional deformation of the slab around 150-km depth along the Nazca Plate subduction direction, associated with large-scale fluid release. Furthermore, the lateral migration of the gravity signal towards the surface from a low coupling segment around North to the high coupling one in the South suggests that the Mw 8.8 earthquake may have originated from the propagation up to the trench of this deeper slab deformation. Our results highlight the importance of observations of the Earth's time-varying gravity field from satellites in order to probe slow mass redistributions in-depth major plate boundaries and provide new information on dynamic processes in the subduction system, essential to better understand the seismic cycle as a whole. Numéro de notice : A2022-280 Affiliation des auteurs : UMR IPGP-Géod+Ext (2020- ) Thématique : POSITIONNEMENT Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.epsl.2022.117465 En ligne : https://doi.org/10.1016/j.epsl.2022.117465 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=100288 [article]

A combination of convolutional and graph neural networks for regularized road surface extraction / Jingjing Yan in IEEE Transactions on geoscience and remote sensing, vol 60 n° 2 (February 2022)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 60 n° 2 (February 2022) . - n° 4409113 Titre : A combination of convolutional and graph neural networks for regularized road surface extraction Type de document : Article/Communication Auteurs : Jingjing Yan, Auteur ; Shunping Ji, Auteur ; Yao Wei, Auteur Année de publication : 2022 Article en page(s) : n° 4409113 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] Bavière (Allemagne) [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] détection de contours [Termes IGN] extraction de traits caractéristiques [Termes IGN] extraction du réseau routier [Termes IGN] image aérienne [Termes IGN] jeu de données [Termes IGN] optimisation (mathématiques) [Termes IGN] régression [Termes IGN] réseau neuronal de graphes [Termes IGN] Wuhan (Chine) Résumé : (auteur) Road surface extraction from high-resolution remote sensing images has many engineering applications; however, extracting regularized and smooth road surface maps that reach the human delineation level is a very challenging task, and substantial and time-consuming manual work is usually unavoidable. In this article, to solve this problem, we propose a novel regularized road surface extraction framework by introducing a graph neural network (GNN) for processing the road graph that is preconstructed from the easily accessible road centerlines. The proposed framework formulates the road surface extraction problem as two-sided width inference of the road graph and consists of a convolutional neural network (CNN)-based feature extractor and a GNN model for vertex attribute adjustment. The CNN extracts the high-level abstract features of each vertex in the graph as the input of the GNN and also the road boundary features that allow us to distinguish roads from the background. The GNN propagates and aggregates the features of the vertices in the graph to achieve global optimization of the regression of the regularized widths of the vertices. At the same time, a biased centerline map can also be corrected based on the width prediction result. To the best of the authors’ knowledge, this is the first study to have introduced a GNN to regularized human-level road surface extraction. The proposed method was evaluated on four diverse datasets, and the results show that the proposed method comprehensively outperforms the recent CNN-based segmentation methods and other regularization methods in the intersection over union (IoU) and smoothness score, and a visual check shows that a majority of the prediction results of the proposed method approach the human delineation level. Numéro de notice : A2022-297 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2022.3151688 Date de publication en ligne : 15/02/2022 En ligne : https://doi.org/10.1109/TGRS.2022.3151688 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=100355 [article]

Evaluation de méthodes automatisées de cartographie des zones inondables adaptées à la prévision des crues soudaines / Nabil Hocini (2022)  

Permalink

Improving the Fagacées growth model with an expanded common beech (Fagus sylvatica L.) data series from France and Germany / Gilles Le Moguédec in Annals of Forest Science [en ligne], vol 78 n° 4 (December 2021)  

Permalink

Orbit error removal in InSAR/MTInSAR with a patch-based polynomial model / Yanan Du in International journal of applied Earth observation and geoinformation, vol 102 (October 2021)  

Permalink

Efficient image dataset classification difficulty estimation for predicting deep-learning accuracy / Florian Scheidegger in The Visual Computer, vol 37 n° 6 (June 2021)  

Permalink

Refinement of interferometric SAR parameters using digital terrain model as an external reference / Jyunpei Uemoto in ISPRS Journal of photogrammetry and remote sensing, vol 175 (May 2021)  

Permalink

Anti-cross validation technique for constructing and boosting random subspace neural network ensembles for hyperspectral image classification / Laxmi Narayana Eeti in Geocarto international, vol 36 n° 6 ([01/04/2021])  

Permalink

A shape transformation-based dataset augmentation framework for pedestrian detection / Zhe Chen in International journal of computer vision, vol 129 n° 4 (April 2021)  

Permalink

Semi-supervised joint learning for hand gesture recognition from a single color image / Chi Xu in Sensors, vol 21 n° 3 (February 2021)  

Permalink

Suivi du statut hydrique de la vigne par télédétection hyper et multispectrale / Eve Laroche (2021)  

Permalink

The challenge of robust trait estimates with deep learning on high resolution RGB images / Etienne David (2021)  

Permalink