Centre de documentation
scientifique

GA-Net: A geometry prior assisted neural network for road extraction / Xin Chen in International journal of applied Earth observation and geoinformation, vol 114 (November 2022)  

Public [article]  inInternational journal of applied Earth observation and geoinformation > vol 114 (November 2022) . - n° 103004 Titre : GA-Net: A geometry prior assisted neural network for road extraction Type de document : Article/Communication Auteurs : Xin Chen, Auteur ; Qun Sun, Auteur ; Wenyue Guo, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : n° 103004 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] attention (apprentissage automatique) [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] détection de contours [Termes IGN] données multiéchelles [Termes IGN] extraction automatique [Termes IGN] extraction de traits caractéristiques [Termes IGN] extraction du réseau routier [Termes IGN] jeu de données [Termes IGN] Massachusetts (Etats-Unis) Résumé : (auteur) With geospatial intelligence research developing rapidly, automatic road extraction is becoming a fundamental and challenging task. Due to the special geometric structure and spectral information of road networks, existing methods suffer from incomplete and fractured results. In this work, a novel road extraction convolutional neural network, incorporating the road boundary details and road junction information via a dual-branch multi-task structure, is proposed to learn synergistic feature representations and strengthen road connectivity. Firstly, a BiFPN-based feature aggregation module is utilised to bridge the semantic gap between low-level and high-level feature maps, allowing multi-scale spatial details to be fully fused. Secondly, the boundary auxiliary branch, using a U-shaped network with a spatial-channel attention module, captures residential information for the backbone to enhance the subtleties of road edges. Thirdly, the node inferring branch models the road junction position jointly with the road surface, aiming to strengthen the topology structure and reduce the fragmented road segments. We perform experiments on three diverse road datasets, namely the DeepGlobe dataset, Massachusetts dataset, and SpaceNet dataset. The results demonstrate that our model shows an overall performance improvement over some SOTA algorithms and the IoU indicator achieves 3.86%, 0.79%, and 1.71% improvements over Unet on the three datasets, respectively. Numéro de notice : A2022-785 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.1016/j.jag.2022.103004 En ligne : https://doi.org/10.1016/j.jag.2022.103004 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=101888 [article]

Predicting user activity intensity using geographic interactions based on social media check-in data / Jing Li in ISPRS International journal of geo-information, vol 10 n° 8 (August 2021)  

Public [article]  inISPRS International journal of geo-information > vol 10 n° 8 (August 2021) . - n° 555 Titre : Predicting user activity intensity using geographic interactions based on social media check-in data Type de document : Article/Communication Auteurs : Jing Li, Auteur ; Wenyue Guo, Auteur ; Haiyan Liu, Auteur ; et al., Auteur Année de publication : 2021 Article en page(s) : n° 555 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Analyse spatiale [Termes IGN] apprentissage profond [Termes IGN] données issues des réseaux sociaux [Termes IGN] interaction spatiale [Termes IGN] mobilité humaine [Termes IGN] modèle non linéaire [Termes IGN] relation spatiale [Termes IGN] réseau neuronal de graphes [Termes IGN] réseau neuronal récurrent [Termes IGN] utilisateur Résumé : (auteur) Predicting user activity intensity is crucial for various applications. However, existing studies have two main problems. First, as user activity intensity is nonstationary and nonlinear, traditional methods can hardly fit the nonlinear spatio-temporal relationships that characterize user mobility. Second, user movements between different areas are valuable, but have not been utilized for the construction of spatial relationships. Therefore, we propose a deep learning model, the geographical interactions-weighted graph convolutional network-gated recurrent unit (GGCN-GRU), which is good at fitting nonlinear spatio-temporal relationships and incorporates users’ geographic interactions to construct spatial relationships in the form of graphs as the input. The model consists of a graph convolutional network (GCN) and a gated recurrent unit (GRU). The GCN, which is efficient at processing graphs, extracts spatial features. These features are then input into the GRU, which extracts their temporal features. Finally, the GRU output is passed through a fully connected layer to obtain the predictions. We validated this model using a social media check-in dataset and found that the geographical interactions graph construction method performs better than the baselines. This indicates that our model is appropriate for fitting the complex nonlinear spatio-temporal relationships that characterize user mobility and helps improve prediction accuracy when considering geographic flows. Numéro de notice : A2021-588 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.3390/ijgi10080555 Date de publication en ligne : 17/08/2021 En ligne : https://doi.org/10.3390/ijgi10080555 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=98206 [article]

Bias compensation for rational function model based on total least squares / Anzhu Yu in Photogrammetric record, vol 32 n° 157 (March - May 2017)  

Public [article]  inPhotogrammetric record > vol 32 n° 157 (March - May 2017) . - pp 48 - 60 Titre : Bias compensation for rational function model based on total least squares Type de document : Article/Communication Auteurs : Anzhu Yu, Auteur ; Ting Jiang, Auteur ; Wenyue Guo, Auteur Année de publication : 2017 Article en page(s) : pp 48 - 60 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Photogrammétrie numérique [Termes IGN] analyse numérique [Termes IGN] calcul d'erreur [Termes IGN] compensation par moindres carrés [Termes IGN] erreur aléatoire [Termes IGN] erreur systématique [Termes IGN] modèle par fonctions rationnelles [Termes IGN] pondération [Termes IGN] précision du positionnement Résumé : (auteur) When using the rational function model for the geometric orientation and geopositioning of satellite imagery, systematic bias compensation for vendor-provided rational polynomial coefficients (RPCs) is very important. Most existing bias-compensation models express systematic biases as a function of certain deterministic parameters, and least squares adjustment is used for estimating correction parameters. In this paper, the errors-in-variables model is introduced to take random errors in both the observation vector and the design matrix into consideration, based on a weighted total least squares adjustment. Experiments performed with two datasets demonstrate that the proposed method is reliable and the geopositioning accuracy improvement is better compared with a traditional least squares adjustment. Numéro de notice : A2017-197 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.1111/phor.12183 En ligne : http://dx.doi.org/10.1111/phor.12183 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=84871 [article]