Centre dedocumentation
scientifique

A hierarchical deep learning framework for the consistent classification of land use objects in geospatial databases / Chun Yang in ISPRS Journal of photogrammetry and remote sensing, vol 177 (July 2021)  

Public [article]  inISPRS Journal of photogrammetry and remote sensing > vol 177 (July 2021) . - pp 38 - 56 Titre : A hierarchical deep learning framework for the consistent classification of land use objects in geospatial databases Type de document : Article/Communication Auteurs : Chun Yang, Auteur ; Franz Rottensteiner, Auteur ; Christian Heipke, Auteur Année de publication : 2021 Article en page(s) : pp 38 - 56 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Bases de données localisées [Termes IGN] Allemagne [Termes IGN] apprentissage profond [Termes IGN] approche hiérarchique [Termes IGN] classification automatique d'objets [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] image aérienne [Termes IGN] jointure [Termes IGN] objet géographique [Termes IGN] occupation du sol [Termes IGN] optimisation (mathématiques) [Termes IGN] utilisation du sol Résumé : (Auteur) Land use as contained in geospatial databases constitutes an essential input for different applications such as urban management, regional planning and environmental monitoring. In this paper, a hierarchical deep learning framework is proposed to verify the land use information. For this purpose, a two-step strategy is applied. First, given high-resolution aerial images, the land cover information is determined. To achieve this, an encoder-decoder based convolutional neural network (CNN) is proposed. Second, the pixel-wise land cover information along with the aerial images serves as input for another CNN to classify land use. Because the object catalogue of geospatial databases is frequently constructed in a hierarchical manner, we propose a new CNN-based method aiming to predict land use in multiple levels hierarchically and simultaneously. A so called Joint Optimization (JO) is proposed where predictions are made by selecting the hierarchical tuple over all levels which has the maximum joint class scores, providing consistent results across the different levels. The conducted experiments show that the CNN relying on JO outperforms previous results, achieving an overall accuracy up to 92.5%. In addition to the individual experiments on two test sites, we investigate whether data showing different characteristics can improve the results of land cover and land use classification, when processed together. To do so, we combine the two datasets and undertake some additional experiments. The results show that adding more data helps both land cover and land use classification, especially the identification of underrepresented categories, despite their different characteristics. Numéro de notice : A2021-370 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE/IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.isprsjprs.2021.04.022 date de publication en ligne : 13/05/2021 En ligne : https://doi.org/10.1016/j.isprsjprs.2021.04.022 Format de la ressource électronique : URL Article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=97774 [article]

Réservation

Réserver ce document

Exemplaires (3)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
081-2021071	SL	Revue	Centre de documentation	Revues en salle	Disponible
081-2021073	DEP-RECP	Revue	LaSTIG	Dépôt en unité	Exclu du prêt
081-2021072	DEP-RECF	Revue	Nancy	Dépôt en unité	Exclu du prêt

3D reconstruction of bridges from airborne laser scanning data and cadastral footprints / Steffen Goebbels in Journal of Geovisualization and Spatial Analysis, vol 5 n° 1 (June 2021)  

Public [article]  inJournal of Geovisualization and Spatial Analysis > vol 5 n° 1 (June 2021) . - n° 10 Titre : 3D reconstruction of bridges from airborne laser scanning data and cadastral footprints Type de document : Article/Communication Auteurs : Steffen Goebbels, Auteur Année de publication : 2021 Article en page(s) : n° 10 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Lasergrammétrie [Termes IGN] Allemagne [Termes IGN] axe médian [Termes IGN] CityGML [Termes IGN] données cadastrales [Termes IGN] données lidar [Termes IGN] empreinte [Termes IGN] pont [Termes IGN] Ransac (algorithme) [Termes IGN] reconstruction 3D [Termes IGN] semis de points Résumé : (auteur) The given paper describes a method for automatic 3D reconstruction of bridges from cadastral footprints and airborne laser scanning point clouds. The reconstructed bridges are used to enrich 3D city models. Unlike roofs, decks of bridges are typically smooth without ridge lines or step edges. Therefore, established methods for roof reconstruction are not suitable for bridges. The standard description language for semantic city models is CityGML. This specification of the Open Geospatial Consortium assumes that surfaces are composed of planar polygons. The approximation of smooth decks by planar polygons is achieved by using a medial axis tree. Instead of the medial axis of the footprint, a modified medial axis is computed that does not consider counter bearing edges. The resulting tree represents centerline connections between all counter bearing edges and, in conjunction with filtered height values of a point cloud, serves as the basis for approximation with polygons. In addition to modeling decks, superstructures such as pylons and cables are also derived from the point cloud. For this purpose, planes carrying many superstructure points are detected using the Random Sampling Consensus Algorithm (RANSAC). Images are generated by projecting points onto these planes. Then, image processing methods are used to find connected contours that are extruded to form 3D objects. The presented method was successfully applied to all bridges of two German cities as well as to large bridges built over the Rhine River. Numéro de notice : A2021-359 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.1007/s41651-021-00076-9 date de publication en ligne : 12/04/2021 En ligne : https://doi.org/10.1007/s41651-021-00076-9 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=97623 [article]

Weak relationships of continuous forest management intensity and remotely sensed stand structural complexity in temperate mountain forests / Thomas Asbeck in European Journal of Forest Research, vol 140 n° 3 (June 2021)  

Public [article]  inEuropean Journal of Forest Research > vol 140 n° 3 (June 2021) . - pp 721 - 731 Titre : Weak relationships of continuous forest management intensity and remotely sensed stand structural complexity in temperate mountain forests Type de document : Article/Communication Auteurs : Thomas Asbeck, Auteur ; Julian Frey, Auteur Année de publication : 2021 Article en page(s) : pp 721 - 731 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Lasergrammétrie [Termes IGN] Abies alba [Termes IGN] composition d'un peuplement forestier [Termes IGN] diamètre à hauteur de poitrine [Termes IGN] données lidar [Termes IGN] données localisées 3D [Termes IGN] Forêt-Noire, massif de la [Termes IGN] gestion forestière durable [Termes IGN] modélisation de la forêt [Termes IGN] Picea abies [Termes IGN] semis de points [Termes IGN] structure d'un peuplement forestier Résumé : (auteur) Understanding the relationship of stand structural complexity and forest management is relevant to create desired stand structures by adapting management strategies under changing disturbance scenarios and climatic conditions. To overcome difficulties in differentiating between strict categories of silvicultural practices and to describe the impact of forest management more appropriate, we used a continuous indicator of forest management intensity (ForMI). The ForMI consists of three components including volumes of natural deadwood, non-native tree species and harvested trees. There are a great number of approaches to quantify stand structure; here we used the recently established stand structural complexity index (SSCI) which represents a density-dependent as well as vertical measure of complexity based on the distribution of points in 3D space inventoried by terrestrial laser scanning. The data collection took place in 135 one-hectare plots managed under close-to-nature forest management (CTNFM) located in the Black Forest, Germany. We build generalized additive models to test the relationship of the SSCI with the ForMI. The model results did not prove a significant relationship between the SSCI and the ForMI, but components of the ForMI showed significant relationships to the SSCI. Our results indicate that the relationship between stand structural complexity and forest management intensity is, while plausible, not trivial to demonstrate. We conclude that forest managers have a relatively wide range of choices in CTNFM to adapt forests within a similar range of management intensity as presented here to future challenges, since management intensity does not change the forest structure drastically. Numéro de notice : A2021-431 Affiliation des auteurs : non IGN Thématique : FORET/IMAGERIE Nature : Article DOI : 10.1007/s10342-021-01361-4 date de publication en ligne : 13/02/2021 En ligne : https://doi.org/10.1007/s10342-021-01361-4 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=97793 [article]

Development of German-Ukrainian cooperations for education and research in photogrammetry and laser scanning / Thomas Luhmann in Geo-spatial Information Science, vol 24 n° 1 (March 2021)  

Public [article]  inGeo-spatial Information Science > vol 24 n° 1 (March 2021) . - pp 152 - 159 Titre : Development of German-Ukrainian cooperations for education and research in photogrammetry and laser scanning Type de document : Article/Communication Auteurs : Thomas Luhmann, Auteur Année de publication : 2021 Article en page(s) : pp 152 - 159 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Formation [Termes IGN] Allemagne [Termes IGN] apprentissage en ligne [Termes IGN] balayage laser [Termes IGN] coopération internationale [Termes IGN] enseignement supérieur [Termes IGN] formation à distance [Termes IGN] modélisation 3D [Termes IGN] photogrammétrie [Termes IGN] recherche scientifique [Termes IGN] traitement de nuage de points [Termes IGN] Ukraine Résumé : (Auteur) Since 2011 an intensive international cooperation has been in place between the Institute for Applied Photogrammetry and Geoinformatics (IAPG) at the Jade University of Applied Sciences in Oldenburg, Germany, and various Ukrainian universities. Following an initial contact by Prof. Gottfried Konecny, the first visit was organized, and was followed by many more. In subsequent years an intensive cooperation was established particularly with the National University for Construction and Architecture (KNUCA) in Kiev. In addition to architects and civil engineers, KNUCA also trains geodesists, geo-information scientists and landscape planners. The cooperation today includes the reciprocal exchange of scientists and students, research projects, courses and cooperation at many other levels. In addition, a commercial company has been established, SPM3D LLC, which now employs more than 14 engineers in the field of 3D acquisition, point cloud processing and modeling. This article summarizes the history of the cooperation and presents the results of associated student projects. In addition, results of joint work on the development of a virtual laser scanner are presented, part of a German-Ukrainian initiative to digitize teaching. Numéro de notice : A2021-299 Affiliation des auteurs : non IGN Thématique : IMAGERIE/SOCIETE NUMERIQUE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1080/10095020.2020.1831891 date de publication en ligne : 13/10/2020 En ligne : https://doi.org/10.1080/10095020.2020.1831891 Format de la ressource électronique : URL Article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=97384 [article]

European beech leads to more bioactive humus forms but stronger mineral soil acidification as Norway spruce and Scots pine – Results of a repeated site assessment after 63 and 82 years of forest conversion in Central Germany / Florian Achilles in Forest ecology and management, vol 483 ([01/03/2021])  

Public [article]  inForest ecology and management > vol 483 [01/03/2021] . - n° 118769 Titre : European beech leads to more bioactive humus forms but stronger mineral soil acidification as Norway spruce and Scots pine – Results of a repeated site assessment after 63 and 82 years of forest conversion in Central Germany Type de document : Article/Communication Auteurs : Florian Achilles, Auteur ; Alexander Tischer, Auteur ; Markus Bernhardt-Römermann, Auteur ; et al., Auteur Année de publication : 2021 Article en page(s) : n° 118769 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Termes IGN] Allemagne [Termes IGN] analyse diachronique [Termes IGN] biomasse forestière [Termes IGN] composition d'un peuplement forestier [Termes IGN] composition des sols [Termes IGN] conversion forestière [Termes IGN] Fagus (genre) [Termes IGN] gestion forestière [Termes IGN] Picea abies [Termes IGN] Pinus sylvestris [Termes IGN] sol acide [Termes IGN] sol forestier [Vedettes matières IGN] Sylviculture Résumé : (auteur) To reduce the effects of extended coniferous monoculture plantations on forest floor and topsoil processes, like amplified acidification or nutrient immobilization in organic layers, small interspersed groups of European beech were planted at the beginning of the 20th century amid large coniferous stands (CS) in Central Germany. Today, these so-called “Green Eyes” (GE) are 82-year old. In our study we focused on two different timelines to investigate the effects of forest conversion on vegetation composition, forest floor and mineral soil properties, encompassing a long-term (>80 year) comparison from tree planting in the 1930s to 2018 and a shorter timeframe nearly spanning 20 years (1999–2018). Since long-term forest conversion experiments (>60 years) across sites sharing a common forest land-use history are scarce, our study allows to contribute to a better evaluation of the long-term effects on changes in soil properties. We linked standard methods assessing soil pH, organic C and total N concentrations as well as stocks as indicators of soil quality changes in forest floor and mineral soil (down to 40 cm depth) to patterns in ground vegetation dynamics (most GE were characterized only by a few herbaceous species). Our results exhibited an effect of forest conversion on the activation in forest floor turnover resulting in increased turnover rates in the GE (GE contain only half of the forest floor OM of CS), hence forming more bioactive humus forms (mull, moder) and a noticeable higher forest floor pH (+0.6 units) compared to CS. The OC translocation from the forest floor into the topsoil is higher under GE (+0.7% OC content), probably building up a stable SOC pool hence contributing to C sequestration. The positive effects of European beech on forest floor quality are related to a stronger acidification of the subsoil during the past 20 years (-0.5 pH units). The base pump effect (uptake of Ca2+, Mg2+, K+ and assimilation into tree biomass in exchange release of H+) in GE led to more acidic conditions (- 0.2 pH units) in the deeper mineral soil compared to CS. This bio-acidification may serve as one reason for subsoil acidification, as observed in our study, pointing out that mineral soil acidification is still an important issue in central European forests. Future restoration of soil base cation pools will still depend on the rate of bio-acidification, on the amount of acidifying air pollutant (and base cation) depositions and on forest management practices (e.g. intensity of biomass removal, soil regeneration-oriented liming). Numéro de notice : A2021-229 Affiliation des auteurs : non IGN Thématique : FORET Nature : Article DOI : 10.1016/j.foreco.2020.118769 date de publication en ligne : 18/11/2020 En ligne : https://doi.org/10.1016/j.foreco.2020.118769 Format de la ressource électronique : url article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=97216 [article]

Gridded population mapping for Germany based on building density, height and type from Earth Observation data using census disaggregation and bottom-up estimates / Franz Schug in Plos one, vol 16 n° 3 (March 2021)  

Permalink

Modelling potential density of natural regeneration of European oak species (Quercus robur L., Quercus petraea (Matt.) Liebl.) depending on the distance to the potential seed source: Methodological approach for modelling dispersal from inventory data at forest enterprise level / Maximilian Axer in Forest ecology and management, vol 482 ([15/02/2021])  

Permalink

Development and analysis of land-use/land-cover spatio-temporal metrics in urban environments: Exploring urban growth patterns and linkages to socio-economic factors / Marta Sapena Moll (2021)  

Permalink

From point clouds to high-fidelity models - advanced methods for image-based 3D reconstruction / Audrey Richard (2021)  

Permalink

Investigation of Sentinel-1 time series for sensitivity to fern vegetation in an European temperate forest / Marlin Mueller (2021)  

Permalink

Unmixing-based Sentinel-2 downscaling for urban land cover mapping / Fei Xu in ISPRS Journal of photogrammetry and remote sensing, vol 171 (January 2021)  

Permalink

Learning from urban form to predict building heights / Nikola Milojevic-Dupont in Plos one, vol 15 n° 12 (December 2020)  

Permalink

Mapping forest tree species in high resolution UAV-based RGB-imagery by means of convolutional neural networks / Felix Schiefer in ISPRS Journal of photogrammetry and remote sensing, vol 170 (December 2020)  

Permalink

Multistrategy ensemble regression for mapping of built-up density and height with Sentinel-2 data / Christian Geiss in ISPRS Journal of photogrammetry and remote sensing, vol 170 (December 2020)  

Permalink

Chloroplast haplotypes of Northern red oak (Quercus rubra L.) stands in Germany suggest their origin from Northeastern Canada / Jeremias Götz in Forests, vol 11 n° 9 (September 2020)  

Permalink