Descripteur
Documents disponibles dans cette catégorie (406)
Ajouter le résultat dans votre panier
Visionner les documents numériques
Affiner la recherche Interroger des sources externes
Etendre la recherche sur niveau(x) vers le bas
Ultrahigh-resolution boreal forest canopy mapping: Combining UAV imagery and photogrammetric point clouds in a deep-learning-based approach / Linyuan Li in International journal of applied Earth observation and geoinformation, vol 107 (March 2022)
[article]
Titre : Ultrahigh-resolution boreal forest canopy mapping: Combining UAV imagery and photogrammetric point clouds in a deep-learning-based approach Type de document : Article/Communication Auteurs : Linyuan Li, Auteur ; Xihan Mu, Auteur ; Francesco Chianucci, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : n° 102686 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications photogrammétriques
[Termes IGN] algorithme SLIC
[Termes IGN] apprentissage profond
[Termes IGN] canopée
[Termes IGN] carte forestière
[Termes IGN] Chine
[Termes IGN] classification par maximum de vraisemblance
[Termes IGN] classification par réseau neuronal convolutif
[Termes IGN] couvert forestier
[Termes IGN] données d'entrainement (apprentissage automatique)
[Termes IGN] données lidar
[Termes IGN] faisceau laser
[Termes IGN] forêt boréale
[Termes IGN] image captée par drone
[Termes IGN] modèle numérique de surface de la canopée
[Termes IGN] modèle numérique de terrain
[Termes IGN] segmentation sémantique
[Termes IGN] semis de points
[Termes IGN] sous-étage
[Termes IGN] structure-from-motionRésumé : (auteur) Accurate wall-to-wall estimation of forest crown cover is critical for a wide range of ecological studies. Notwithstanding the increasing use of UAVs in forest canopy mapping, the ultrahigh-resolution UAV imagery requires an appropriate procedure to separate the contribution of understorey from overstorey vegetation, which is complicated by the spectral similarity between the two forest components and the illumination environment. In this study, we investigated the integration of deep learning and the combined data of imagery and photogrammetric point clouds for boreal forest canopy mapping. The procedure enables the automatic creation of training sets of tree crown (overstorey) and background (understorey) data via the combination of UAV images and their associated photogrammetric point clouds and expands the applicability of deep learning models with self-supervision. Based on the UAV images with different overlap levels of 12 conifer forest plots that are categorized into “I”, “II” and “III” complexity levels according to illumination environment, we compared the self-supervised deep learning-predicted canopy maps from original images with manual delineation data and found an average intersection of union (IoU) larger than 0.9 for “complexity I” and “complexity II” plots and larger than 0.75 for “complexity III” plots. The proposed method was then compared with three classical image segmentation methods (i.e., maximum likelihood, Kmeans, and Otsu) in the plot-level crown cover estimation, showing outperformance in overstorey canopy extraction against other methods. The proposed method was also validated against wall-to-wall and pointwise crown cover estimates using UAV LiDAR and in situ digital cover photography (DCP) benchmarking methods. The results showed that the model-predicted crown cover was in line with the UAV LiDAR method (RMSE of 0.06) and deviate from the DCP method (RMSE of 0.18). We subsequently compared the new method and the commonly used UAV structure-from-motion (SfM) method at varying forward and lateral overlaps over all plots and a rugged terrain region, yielding results showing that the method-predicted crown cover was relatively insensitive to varying overlap (largest bias of less than 0.15), whereas the UAV SfM-estimated crown cover was seriously affected by overlap and decreased with decreasing overlap. In addition, canopy mapping over rugged terrain verified the merits of the new method, with no need for a detailed digital terrain model (DTM). The new method is recommended to be used in various image overlaps, illuminations, and terrains due to its robustness and high accuracy. This study offers opportunities to promote forest ecological applications (e.g., leaf area index estimation) and sustainable management (e.g., deforestation). Numéro de notice : A2022-192 Affiliation des auteurs : non IGN Thématique : FORET/IMAGERIE Nature : Article DOI : 10.1016/j.jag.2022.102686 Date de publication en ligne : 05/02/2022 En ligne : https://doi.org/10.1016/j.jag.2022.102686 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=99951
in International journal of applied Earth observation and geoinformation > vol 107 (March 2022) . - n° 102686[article]Using street view images to identify road noise barriers with ensemble classification model and geospatial analysis / Kai Zhang in Sustainable Cities and Society, vol 78 (March 2022)
[article]
Titre : Using street view images to identify road noise barriers with ensemble classification model and geospatial analysis Type de document : Article/Communication Auteurs : Kai Zhang, Auteur ; Zhen Qian, Auteur ; Yue Yang, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : n° 103598 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Analyse spatiale
[Termes IGN] analyse de groupement
[Termes IGN] apprentissage profond
[Termes IGN] cartographie du bruit
[Termes IGN] Chine
[Termes IGN] classification par réseau neuronal convolutif
[Termes IGN] distribution spatiale
[Termes IGN] image Streetview
[Termes IGN] lutte contre le bruit
[Termes IGN] milieu urbain
[Termes IGN] OpenStreetMap
[Termes IGN] planification urbaine
[Termes IGN] pollution acoustique
[Termes IGN] trafic routier
[Termes IGN] ville durableRésumé : (auteur) Road noise barriers (RNBs) are important urban infrastructures to relieve the harm of traffic noise pollution for citizens. Therefore, obtaining the spatial distribution characteristics of RNBs, such as precise positions and mileage, can be of great help for obtaining more accurate urban noise maps and assessing the quality of the urban living environment for sustainable urban development. However, an effective and efficient method for identifying RNBs and acquiring their attributes in large areas is scarce. This study constructs an ensemble classification model (ECM) to automatically identify RNBs at the city level based on Baidu Street View (BSV). Firstly, the bootstrap sampling method is proposed to build a street view image-based train set, where the effect of imbalanced categories of samples was reduced by adding confusing negative samples. Secondly, two state-of-the-art deep learning models, ResNet and DenseNet, are ensembled to construct an ECM based on the bagging framework. Finally, a post-processing method has been proposed based on geospatial analysis to eliminate street view images (SVIs) that are misclassified as RNBs. This study takes Suzhou, China as the study area to validate the proposed method. The model achieved an accuracy and F1-score of 0.98 and 0.90, respectively. The total mileage of the RNBs in Suzhou was 178,919 m. The results demonstrated the performance of the proposed RNBs identification framework. The significance of obtaining RNBs attributes for accelerating sustainable urban development has been demonstrated through the case of photovoltaic noise barriers (PVNBs). Numéro de notice : A2022-241 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE/IMAGERIE/INFORMATIQUE Nature : Article DOI : 10.1016/j.scs.2021.103598 Date de publication en ligne : 20/12/2021 En ligne : https://doi.org/10.1016/j.scs.2021.103598 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=100167
in Sustainable Cities and Society > vol 78 (March 2022) . - n° 103598[article]Visual vs internal attention mechanisms in deep neural networks for image classification and object detection / Abraham Montoya Obeso in Pattern recognition, vol 123 (March 2022)
[article]
Titre : Visual vs internal attention mechanisms in deep neural networks for image classification and object detection Type de document : Article/Communication Auteurs : Abraham Montoya Obeso, Auteur ; Jenny Benois-Pineau, Auteur ; Mireya S. García Vázquez, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : n° 108411 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique
[Termes IGN] analyse visuelle
[Termes IGN] apprentissage profond
[Termes IGN] attention (apprentissage automatique)
[Termes IGN] classification par réseau neuronal convolutif
[Termes IGN] détection d'objet
[Termes IGN] données d'entrainement (apprentissage automatique)
[Termes IGN] extraction de traits caractéristiques
[Termes IGN] oculométrie
[Termes IGN] saillance
[Termes IGN] segmentation sémantique
[Termes IGN] visualisation de donnéesRésumé : (auteur) The so-called “attention mechanisms” in Deep Neural Networks (DNNs) denote an automatic adaptation of DNNs to capture representative features given a specific classification task and related data. Such attention mechanisms perform both globally by reinforcing feature channels and locally by stressing features in each feature map. Channel and feature importance are learnt in the global end-to-end DNNs training process. In this paper, we present a study and propose a method with a different approach, adding supplementary visual data next to training images. We use human visual attention maps obtained independently with psycho-visual experiments, both in task-driven or in free viewing conditions, or powerful models for prediction of visual attention maps. We add visual attention maps as new data alongside images, thus introducing human visual attention into the DNNs training and compare it with both global and local automatic attention mechanisms. Experimental results show that known attention mechanisms in DNNs work pretty much as human visual attention, but still the proposed approach allows a faster convergence and better performance in image classification tasks. Numéro de notice : A2022-197 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.1016/j.patcog.2021.108411 Date de publication en ligne : 12/11/2021 En ligne : https://doi.org/10.1016/j.patcog.2021.108411 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=99988
in Pattern recognition > vol 123 (March 2022) . - n° 108411[article]Aboveground biomass estimation of an agro-pastoral ecology in semi-arid Bundelkhand region of India from Landsat data: a comparison of support vector machine and traditional regression models / Dibyendu Deb in Geocarto international, vol 37 n° 4 ([15/02/2022])
[article]
Titre : Aboveground biomass estimation of an agro-pastoral ecology in semi-arid Bundelkhand region of India from Landsat data: a comparison of support vector machine and traditional regression models Type de document : Article/Communication Auteurs : Dibyendu Deb, Auteur ; Shovik Deb, Auteur ; Debasis Chakraborty, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : pp 1043 - 1058 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection
[Termes IGN] biomasse aérienne
[Termes IGN] distribution spatiale
[Termes IGN] image Landsat-8
[Termes IGN] Inde
[Termes IGN] indice de végétation
[Termes IGN] modèle de régression
[Termes IGN] point d'appui
[Termes IGN] régression linéaire
[Termes IGN] régression multiple
[Termes IGN] séparateur à vaste marge
[Termes IGN] zone semi-arideRésumé : (auteur) This study compared the traditional regression models and support vector machine (SVM) for estimation of aboveground biomass (ABG) of an agro-pastoral ecology using vegetation indices derived from Landsat 8 satellite data as explanatory variables . The area falls in the Shivpuri Tehsil of Madhya Pradesh, India, which is predominantly a semi-arid tract of the Bundelkhand region. The Enhanced Vegetation Index-1 (EVI-1) was identified as the most suitable input variable for the regression models, although the collective effect of a number of the vegetation indices was evident. The EVI-1 was also the most suitable input variable to SVM, due to its capacity to distinctly differentiate diverse vegetation classes. The performance of SVM was better over regression models for estimation of the AGB. Based on the SVM-derived and the ground observations, the AGB of the area was precisely mapped for croplands, grassland and rangelands over the entire region. Numéro de notice : A2022-394 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1080/10106049.2020.1756461 Date de publication en ligne : 29/04/2020 En ligne : https://doi.org/10.1080/10106049.2020.1756461 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=100688
in Geocarto international > vol 37 n° 4 [15/02/2022] . - pp 1043 - 1058[article]A method of vision aided GNSS positioning using semantic information in complex urban environment / Rui Zhai in Remote sensing, vol 14 n° 4 (February-2 2022)
[article]
Titre : A method of vision aided GNSS positioning using semantic information in complex urban environment Type de document : Article/Communication Auteurs : Rui Zhai, Auteur ; Yunbin Yuan, Auteur Année de publication : 2022 Article en page(s) : n° 869 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Navigation et positionnement
[Termes IGN] apprentissage profond
[Termes IGN] cartographie et localisation simultanées
[Termes IGN] centrale inertielle
[Termes IGN] filtre de Kalman
[Termes IGN] GNSS assisté pour la navigation
[Termes IGN] information sémantique
[Termes IGN] milieu urbain
[Termes IGN] positionnement cinématique en temps réel
[Termes IGN] positionnement par GNSS
[Termes IGN] segmentation sémantique
[Termes IGN] système de numérisation mobile
[Termes IGN] vision par ordinateurRésumé : (auteur) High-precision localization through multi-sensor fusion has become a popular research direction in unmanned driving. However, most previous studies have performed optimally only in open-sky conditions; therefore, high-precision localization in complex urban environments required an urgent solution. The complex urban environments employed in this study include dynamic environments, which result in limited visual localization performance, and highly occluded environments, which yield limited global navigation satellite system (GNSS) performance. In order to provide high-precision localization in these environments, we propose a vision-aided GNSS positioning method using semantic information by integrating stereo cameras and GNSS into a loosely coupled navigation system. To suppress the effect of dynamic objects on visual positioning accuracy, we propose a dynamic-simultaneous localization and mapping (Dynamic-SLAM) algorithm to extract semantic information from images using a deep learning framework. For the GPS-challenged environment, we propose a semantic-based dynamic adaptive Kalman filtering fusion (S-AKF) algorithm to develop vision aided GNSS and achieve stable and high-precision positioning. Experiments were carried out in GNSS-challenged environments using the open-source KITTI dataset to evaluate the performance of the proposed algorithm. The results indicate that the dynamic-SLAM algorithm improved the performance of the visual localization algorithm and effectively suppressed the error spread of the visual localization algorithm. Additionally, after vision was integrated, the loosely-coupled navigation system achieved continuous high-accuracy positioning in GNSS-challenged environments. Numéro de notice : A2022-167 Affiliation des auteurs : non IGN Thématique : IMAGERIE/POSITIONNEMENT Nature : Article DOI : 10.3390/rs14040869 Date de publication en ligne : 11/02/2022 En ligne : https://doi.org/10.3390/rs14040869 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=99792
in Remote sensing > vol 14 n° 4 (February-2 2022) . - n° 869[article]Simulating fire-safe cities using a machine learning-based algorithm for the complex urban forms of developing nations: a case of Mumbai India / Vaibhav Kumar in Geocarto international, vol 37 n° 4 ([15/02/2022])PermalinkSuspended sediment prediction using integrative soft computing models: on the analogy between the butterfly optimization and genetic algorithms / Marzieh Fadaee in Geocarto international, vol 37 n° 4 ([15/02/2022])PermalinkApprendre à combiner l'information géographique pour générer une carte généralisée [poster à l'EGC 2022] / Azelle Courtial in Revue des Nouvelles Technologies de l'Information, E.38 (2022)PermalinkBuilding footprint extraction in Yangon city from monocular optical satellite image using deep learning / Hein Thura Aung in Geocarto international, vol 37 n° 3 ([01/02/2022])PermalinkA 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)PermalinkDetection of damaged buildings after an earthquake with convolutional neural networks in conjunction with image segmentation / Ramazan Unlu in The Visual Computer, vol 38 n° 2 (February 2022)PermalinkDynamic modelling of rice leaf area index with quad-source optical imagery and machine learning regression models / Lamin R. Mansaray in Geocarto international, vol 37 n° 3 ([01/02/2022])PermalinkEmerging technologies for smart cities’ transportation: Geo-information, data analytics and machine learning approaches / Li-Minn Ang in ISPRS International journal of geo-information, vol 11 n° 2 (February 2022)PermalinkExploring the advantages of the maximum entropy model in calibrating cellular automata for urban growth simulation: a comparative study of four methods / Bin Zhang in GIScience and remote sensing, vol 59 n° 1 (2022)PermalinkGazPNE: annotation-free deep learning for place name extraction from microblogs leveraging gazetteer and synthetic data by rules / Xuke Hu in International journal of geographical information science IJGIS, vol 36 n° 2 (February 2022)Permalink