Centre dedocumentation
scientifique

Exploring multiscale object-based convolutional neural network (multi-OCNN) for remote sensing image classification at high spatial resolution / Vitor Martins in ISPRS Journal of photogrammetry and remote sensing, vol 168 (October 2020)  

Public [article]  inISPRS Journal of photogrammetry and remote sensing > vol 168 (October 2020) . - pp 56 - 73 Titre : Exploring multiscale object-based convolutional neural network (multi-OCNN) for remote sensing image classification at high spatial resolution Type de document : Article/Communication Auteurs : Vitor Martins, Auteur ; Amy L. Kaleita, Auteur ; Brian K. Gelder, Auteur ; et al., Auteur Année de publication : 2020 Article en page(s) : pp 56 - 73 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] analyse d'image orientée objet [Termes IGN] apprentissage profond [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] données multiéchelles [Termes IGN] étiquetage sémantique [Termes IGN] hétérogénéité environnementale [Termes IGN] image à haute résolution [Termes IGN] occupation du sol [Termes IGN] reconnaissance d'objets [Termes IGN] segmentation d'image [Termes IGN] squelettisation Résumé : (auteur) Convolutional Neural Network (CNN) has been increasingly used for land cover mapping of remotely sensed imagery. However, large-area classification using traditional CNN is computationally expensive and produces coarse maps using a sliding window approach. To address this problem, object-based CNN (OCNN) becomes an alternative solution to improve classification performance. However, previous studies were mainly focused on urban areas or small scenes, and implementation of OCNN method is still needed for large-area classification over heterogeneous landscape. Additionally, the massive labeling of segmented objects requires a practical approach for less computation, including object analysis and multiple CNNs. This study presents a new multiscale OCNN (multi-OCNN) framework for large-scale land cover classification at 1-m resolution over 145,740 km2. Our approach consists of three main steps: (i) image segmentation, (ii) object analysis with skeleton-based algorithm, and (iii) application of multiple CNNs for final classification. Also, we developed a large benchmark dataset, called IowaNet, with 1 million labeled images and 10 classes. In our approach, multiscale CNNs were trained to capture the best contextual information during the semantic labeling of objects. Meanwhile, skeletonization algorithm provided morphological representation (“medial axis”) of objects to support the selection of convolutional locations for CNN predictions. In general, proposed multi-OCNN presented better classification accuracy (overall accuracy ~87.2%) compared to traditional patch-based CNN (81.6%) and fixed-input OCNN (82%). In addition, the results showed that this framework is 8.1 and 111.5 times faster than traditional pixel-wise CNN16 or CNN256, respectively. Multiple CNNs and object analysis have proved to be essential for accurate and fast classification. While multi-OCNN produced a high-level of spatial details in the land cover product, misclassification was observed for some classes, such as road versus buildings or shadow versus lake. Despite these minor drawbacks, our results also demonstrated the benefits of IowaNet training dataset in the model performance; overfitting process reduces as the number of samples increases. The limitations of multi-OCNN are partially explained by segmentation quality and limited number of spectral bands in the aerial data. With the advance of deep learning methods, this study supports the claim of multi-OCNN benefits for operational large-scale land cover product at 1-m resolution. Numéro de notice : A2020-634 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.isprsjprs.2020.08.004 date de publication en ligne : 13/08/2020 En ligne : https://doi.org/10.1016/j.isprsjprs.2020.08.004 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=96057 [article]

Réservation

Réserver ce document

Exemplaires (3)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
081-2020101	SL	Revue	Centre de documentation	Revues en salle	Disponible
081-2020103	DEP-RECP	Revue	LaSTIG	Dépôt en unité	Exclu du prêt
081-2020102	DEP-RECF	Revue	Nancy	Dépôt en unité	Exclu du prêt

A comprehensive framework for studying diffusion patterns of imported dengue with individual-based movement data / Haiyan Tao in International journal of geographical information science IJGIS, vol 34 n° 3 (March 2020)  

Public [article]  inInternational journal of geographical information science IJGIS > vol 34 n° 3 (March 2020) . - pp 604 - 624 Titre : A comprehensive framework for studying diffusion patterns of imported dengue with individual-based movement data Type de document : Article/Communication Auteurs : Haiyan Tao, Auteur ; Keli Wang, Auteur ; Li Zhuo, Auteur Année de publication : 2020 Article en page(s) : pp 604 - 624 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Analyse spatiale [Termes IGN] analyse spatio-temporelle [Termes IGN] Chine [Termes IGN] diffusion spatiale [Termes IGN] distribution de Poisson [Termes IGN] données socio-économiques [Termes IGN] hétérogénéité environnementale [Termes IGN] hétérogénéité spatiale [Termes IGN] maladie infectieuse [Termes IGN] migration humaine [Termes IGN] mobilité territoriale [Termes IGN] modèle de régression [Termes IGN] modèle mathématique [Termes IGN] origine - destination [Termes IGN] point d'intérêt [Termes IGN] risque sanitaire [Termes IGN] urbanisation Résumé : (auteur) International communication and global cooperation have greatly accelerated the worldwide spread of dengue fever, increasing the impact of imported cases on dengue outbreaks in non-naturally endemic areas. Existing studies mostly focus on describing the quantitative relationship between imported cases and local transmission but ignore the space-time diffusion mode of imported cases under the influence of individual mobility. In this paper, we propose a comprehensive framework at a fine scale to establish the disease transmission network and a mathematical model, which constructs ‘source-sink’ links between the imported and indigenous cases on a regular grid with a spatial resolution of 1 km to explore the diffusion pattern and spatiotemporal heterogeneity of imported cases. An application to Guangzhou, China, reveals the main flow and transmission path of imported cases under the influence of human movement and identifies the spatiotemporal distribution of transmission speed according to the time lag of each source-sink link. In addition, we demonstrate that using individual-based movement data and socio-economic factors to study human mobility and imported cases can help to understand the driving forces of dengue spread. Our research provides a comprehensive framework for the analysis of early dengue transmission patterns with benefits to similar urban applications. Numéro de notice : A2020-107 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE/URBANISME Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1080/13658816.2019.1684497 date de publication en ligne : 18/11/2019 En ligne : https://doi.org/10.1080/13658816.2019.1684497 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=94707 [article]

Réservation

Réserver ce document

Exemplaires (1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
079-2020031	SL	Revue	Centre de documentation	Revues en salle	Disponible

Patterns of tree diameter distributions in managed and unmanaged Abies alba Mill. and Fagus sylvatica L. forest patches / Rafał Podlaski in Forest ecology and management, vol 435 (1 March 2019)  

Public [article]  inForest ecology and management > vol 435 (1 March 2019) . - pp 7 - 105 Titre : Patterns of tree diameter distributions in managed and unmanaged Abies alba Mill. and Fagus sylvatica L. forest patches Type de document : Article/Communication Auteurs : Rafał Podlaski, Auteur ; Tomasz Sobala, Auteur ; Maciej Kocurek, Auteur Année de publication : 2019 Article en page(s) : pp 7 - 105 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Végétation [Termes IGN] Abies alba [Termes IGN] diamètre à hauteur de poitrine [Termes IGN] diamètre des arbres [Termes IGN] distribution spatiale [Termes IGN] Europe centrale [Termes IGN] Fagus sylvatica [Termes IGN] forêt tempérée [Termes IGN] hétérogénéité environnementale [Termes IGN] phytogéographie Résumé : (Auteur) Temperate forests with shade-tolerant canopy tree species can develop vertical structures of varying complexity. Forests with Abies alba Mill. and Fagus sylvatica L. can be composed of one-, two-, and multi-storied patches and selection patches. A dominant view in forest ecology is that unmanaged forests tend to have greater structural heterogeneity than managed stands. Structural integrity, however, may differ among forest developmental stages. The main objective of this study was to compare the tree diameter complexity in managed and unmanaged patches during the early developmental stage. Data were collected between 2016 and 2018 in the Świętokrzyskie Mountains in Central Europe. The investigated tree communities were dominated by A. alba and F. sylvatica. Sample plots representing the growing-up developmental stage were randomly selected; of these, 30 plots were in managed stands, and 30 plots were in unmanaged forests. The diameter at breast height (DBH) distribution patterns were determined using hierarchical cluster analysis (HCA), clustering indices, and finite mixture models. Three main DBH distribution patterns were identified for the managed stands (K-A, K-B, and K-C). These patterns consisted of three or two sub-populations. The patterns represented structurally diversified patches composed of trees of all ages with multi-, three- or two-layered canopies and with intensive natural processes of regeneration. Two main DBH distribution patterns were identified for the unmanaged forests (S-A, and S-B). These patterns consisted of two clearly separated sub-populations. They are typical in patches with two-layered canopies, and the trees from the upper layer had a large share (40–60%). The distinguished DBH distribution patterns indicated there was greater tree size diversity in the managed stands than in the unmanaged forests. When comparing managed versus unmanaged patches, it is important to consider the developmental stage. Numéro de notice : A2019-185 Affiliation des auteurs : non IGN Thématique : FORET/MATHEMATIQUE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.foreco.2018.12.046 date de publication en ligne : 04/01/2019 En ligne : https://doi.org/10.1016/j.foreco.2018.12.046 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=92719 [article]

Effect of forest structure on stand productivity in Central European forests depends on developmental stage and tree species diversity / Laura Zeller in Forest ecology and management, vol 434 (28 February 2019)  

Public [article]  inForest ecology and management > vol 434 (28 February 2019) . - pp 193 - 204 Titre : Effect of forest structure on stand productivity in Central European forests depends on developmental stage and tree species diversity Type de document : Article/Communication Auteurs : Laura Zeller, Auteur ; Hans Pretzsch, Auteur Année de publication : 2019 Article en page(s) : pp 193 - 204 Note générale : bibliography Langues : Anglais (eng) Descripteur : [Termes IGN] biodiversité végétale [Termes IGN] classe d'âge [Termes IGN] coefficient de Gini [Termes IGN] croissance végétale [Termes IGN] écosystème forestier [Termes IGN] entropie de Shannon [Termes IGN] Europe centrale [Termes IGN] gestion forestière [Termes IGN] hétérogénéité environnementale [Termes IGN] peuplement forestier [Termes IGN] productivité [Termes IGN] structure d'un peuplement forestier [Termes IGN] volume en bois [Vedettes matières IGN] Sylviculture Résumé : (auteur) Recently, many studies have found positive biodiversity–productivity relationships in forests. In contrast, different types of correlations have been identified in the analyses of tree diversity–structure–productivity relationships. We suspect that these conflicting conclusions might result from the different developmental stages of the investigated forest stands. We therefore analyzed the development of tree diversity–structure–productivity relationships at the stand level and individual tree level in 192 long-term experimental plots in Central Europe. As a measure of stand productivity, we analyzed stand volume growth (m3 ha−1 year−1). Tree species diversity was quantified by the Shannon index and structural heterogeneity was represented by the Gini coefficient of basal area. For a more detailed analysis at the tree level using a smaller portion of the dataset, the tree position–dependent indices, diameter differentiation index, and aggregation index were used. Whether the effect of structural heterogeneity on stand productivity was positive or negative depended on the stand development stage. In early developmental stages, high structural heterogeneity lowered productivity. In later developmental stages, however, stand structural heterogeneity had a positive effect on productivity. Our study might provide insights regarding the mechanisms underlying the contradictory findings obtained in recent studies dealing with tree diversity–structure–productivity relationships. This knowledge is vital for the adaptation of forest management to meet future demands on forest ecosystems. Numéro de notice : A2019-011 Affiliation des auteurs : non IGN Thématique : FORET Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.foreco.2018.12.024 date de publication en ligne : 19/12/2018 En ligne : https://doi.org/10.1016/j.foreco.2018.12.024 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=91613 [article]

A simple approach to forest structure classification using airborne laser scanning that can be adopted across bioregions / Syed Adnan in Forest ecology and management, vol 433 (15 February 2019)  

Public [article]  inForest ecology and management > vol 433 (15 February 2019) . - pp 111 - 121 Titre : A simple approach to forest structure classification using airborne laser scanning that can be adopted across bioregions Type de document : Article/Communication Auteurs : Syed Adnan, Auteur ; Matti Maltamo, Auteur ; David A. Coomes, Auteur ; Antonio Garcia-Abril, Auteur ; et al., Auteur Année de publication : 2019 Article en page(s) : pp 111 - 121 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Lasergrammétrie [Termes IGN] canopée [Termes IGN] classification ascendante hiérarchique [Termes IGN] classification barycentrique [Termes IGN] classification et arbre de régression [Termes IGN] coefficient de Gini [Termes IGN] diamètre à hauteur de poitrine [Termes IGN] données lidar [Termes IGN] données localisées 3D [Termes IGN] dynamique de la végétation [Termes IGN] écorégion [Termes IGN] hétérogénéité environnementale [Termes IGN] inventaire forestier étranger (données) [Termes IGN] Pinophyta [Termes IGN] structure d'un peuplement forestier Résumé : (auteur) Reliable assessment of forest structural types (FSTs) aids sustainable forest management. We developed a methodology for the identification of FSTs using airborne laser scanning (ALS), and demonstrate its generality by applying it to forests from Boreal, Mediterranean and Atlantic biogeographical regions. First, hierarchal clustering analysis (HCA) was applied and clusters (FSTs) were determined in coniferous and deciduous forests using four forest structural variables obtained from forest inventory data – quadratic mean diameter , Gini coefficient , basal area larger than mean and density of stems –. Then, classification and regression tree analysis (CART) were used to extract the empirical threshold values for discriminating those clusters. Based on the classification trees, and were the most important variables in the identification of FSTs. Lower, medium and high values of and characterize single storey FSTs, multi-layered FSTs and exponentially decreasing size distributions (reversed J), respectively. Within each of these main FST groups, we also identified young/mature and sparse/dense subtypes using and . Then we used similar structural predictors derived from ALS – maximum height (), L-coefficient of variation (), L-skewness (), and percentage of penetration (), – and a nearest neighbour method to predict the FSTs. We obtained a greater overall accuracy in deciduous forest (0.87) as compared to the coniferous forest (0.72). Our methodology proves the usefulness of ALS data for structural heterogeneity assessment of forests across biogeographical regions. Our simple two-tier approach to FST classification paves the way toward transnational assessments of forest structure across bioregions. Numéro de notice : A2019-007 Affiliation des auteurs : non IGN Thématique : FORET/IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.foreco.2018.10.057 date de publication en ligne : 03/11/2018 En ligne : https://doi.org/10.1016/j.foreco.2018.10.057 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=91600 [article]

Remote Sensing of Evapotranspiration / Prasanna Gowda (2019)  

Permalink

Within- and between-tree variation of wood density components in Pinus nigra at six sites in Portugal / Alexandra Dias in Annals of Forest Science [en ligne], vol 75 n° 2 (June 2018)  

Permalink

Plant species coexistence at local scale in temperate swamp forest: test of habitat heterogeneity hypothesis / Jan Douda in Oecologia, vol 169 n° 2 (June 2012)

Permalink