Descripteur
Documents disponibles dans cette catégorie (1332)
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
Dirichlet process based active learning and discovery of unknown classes for hyperspectral image classification / Hao Wu in IEEE Transactions on geoscience and remote sensing, vol 54 n° 8 (August 2016)
[article]
Titre : Dirichlet process based active learning and discovery of unknown classes for hyperspectral image classification Type de document : Article/Communication Auteurs : Hao Wu, Auteur ; Saurabh Prasad, Auteur Année de publication : 2016 Article en page(s) : pp 4882 - 4895 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique
[Termes IGN] apprentissage automatique
[Termes IGN] classification
[Termes IGN] image hyperspectrale
[Termes IGN] problème de DirichletRésumé : (Auteur) Active learning is an area of significant ongoing research interest for the classification of remotely sensed data, where obtaining efficient training data is both time consuming and expensive. The goal of active learning is to achieve high classification performance by querying as few samples as possible from a large unlabeled data pool. Traditional active learning frameworks all assume the existence of labeled samples for all classes of interest. However, in real-world applications, the unlabeled data pool may contain data from unknown classes that we are not aware of in advance, and a quick detection of them is useful for enriching our training set. In this scenario, traditional active learning methods may not effectively and rapidly detect the unknown classes. We proposed an active learning framework which provides robust classification performance with minimum manual labeling effort while simultaneously discovering unknown (missing) classes. The discovery of unknown classes is particularly suited to an active learning framework where an annotator is in the loop. A Dirichlet process mixture model is utilized in our proposed method to cluster the labeled and unlabeled samples as a whole. If unknown classes exist, they will emerge as new clusters which are different from other existing clusters occupied by known classes, and then, the proposed query strategy will give priority to querying samples in the new clusters. We present experimental results with hyperspectral data to show that our method provides better classification performance compared to existing active learning methods with or without unknown classes. Numéro de notice : A2016-892 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2016.2552507 En ligne : http://dx.doi.org/10.1109/TGRS.2016.2552507 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=83072
in IEEE Transactions on geoscience and remote sensing > vol 54 n° 8 (August 2016) . - pp 4882 - 4895[article]Disaggregation of remotely sensed soil moisture in heterogeneous landscapes using holistic structure-based models / Subit Chakrabarti in IEEE Transactions on geoscience and remote sensing, vol 54 n° 8 (August 2016)
[article]
Titre : Disaggregation of remotely sensed soil moisture in heterogeneous landscapes using holistic structure-based models Type de document : Article/Communication Auteurs : Subit Chakrabarti, Auteur ; Jasmeet Judge, Auteur ; Tara Bongiovanni, Auteur ; et al., Auteur Année de publication : 2016 Article en page(s) : pp 4629 - 4641 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique
[Termes IGN] algorithme d'apprentissage
[Termes IGN] cultures
[Termes IGN] désagrégation
[Termes IGN] Floride (Etats-Unis)
[Termes IGN] humidité du sol
[Termes IGN] modèle de régressionRésumé : (Auteur) In this paper, a novel machine learning algorithm is presented for disaggregation of satellite soil moisture (SM) based on self-regularized regressive models (SRRMs) using high-resolution correlated information from auxiliary sources. It includes regularized clustering that assigns soft memberships to each pixel at a fine scale followed by a kernel regression that computes the value of the desired variable at all pixels. Coarse-scale remotely sensed SM was disaggregated from 10 to 1 km using land cover (LC), precipitation, land surface temperature, leaf area index, and in situ observations of SM. This algorithm was evaluated using multiscale synthetic observations in NC Florida for heterogeneous agricultural LCs. It was found that the rmse for 96% of the pixels was less than 0.02 m 3/m3. The clusters generated represented the data well and reduced the rmse by up to 40% during periods of high heterogeneity in LC and meteorological conditions. The Kullback-Leibler divergence (KLD) between the true SM and the disaggregated estimates is close to zero, for both vegetated and bare-soil LCs. The disaggregated estimates were compared with those generated by the principle of relevant information (PRI) method. The rmse for the PRI disaggregated estimates is higher than the rmse for the SRRM on each day of the season. The KLD of the disaggregated estimates generated by the SRRM is at least four orders of magnitude lower than those for the PRI disaggregated estimates, whereas the computational time needed was reduced by three times. The results indicate that the SRRM can be used for disaggregating SM with complex nonlinear correlations on a grid with high accuracy. Numéro de notice : A2016-888 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2016.2547389 En ligne : http://dx.doi.org/10.1109/TGRS.2016.2547389 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=83068
in IEEE Transactions on geoscience and remote sensing > vol 54 n° 8 (August 2016) . - pp 4629 - 4641[article]Sea ice concentration estimation during melt from dual-pol SAR scenes using deep convolutional neural networks: a case study / Lei Wang in IEEE Transactions on geoscience and remote sensing, vol 54 n° 8 (August 2016)
[article]
Titre : Sea ice concentration estimation during melt from dual-pol SAR scenes using deep convolutional neural networks: a case study Type de document : Article/Communication Auteurs : Lei Wang, Auteur ; K. Andrea Scott, Auteur ; Linlin Xu, Auteur ; David A. Clausi, Auteur Année de publication : 2016 Article en page(s) : pp 4524 - 4533 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image radar et applications
[Termes IGN] classification par réseau neuronal
[Termes IGN] eau de fonte
[Termes IGN] glace de mer
[Termes IGN] iceberg
[Termes IGN] image Radarsat
[Termes IGN] navigation maritime
[Termes IGN] réseau neuronal convolutifRésumé : (Auteur) High-resolution ice concentration maps are of great interest for ship navigation and ice hazard forecasting. In this case study, a convolutional neural network (CNN) has been used to estimate ice concentration using synthetic aperture radar (SAR) scenes captured during the melt season. These dual-pol RADARSAT-2 satellite images are used as input, and the ice concentration is the direct output from the CNN. With no feature extraction or segmentation postprocessing, the absolute mean errors of the generated ice concentration maps are less than 10% on average when compared with manual interpretation of the ice state by ice experts. The CNN is demonstrated to produce ice concentration maps with more detail than produced operationally. Reasonable ice concentration estimations are made in melt regions and in regions of low ice concentration. Numéro de notice : A2016-886 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2016.2543660 En ligne : https://doi.org/10.1109/TGRS.2016.2543660 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=83066
in IEEE Transactions on geoscience and remote sensing > vol 54 n° 8 (August 2016) . - pp 4524 - 4533[article]Unsupervised classification of airborne laser scanning data to locate potential wildlife habitats for forest management planning / Jari Vauhkonen in Forestry, an international journal of forest research, vol 89 n° 4 (August 2016)
[article]
Titre : Unsupervised classification of airborne laser scanning data to locate potential wildlife habitats for forest management planning Type de document : Article/Communication Auteurs : Jari Vauhkonen, Auteur ; Joni Imponen, Auteur Année de publication : 2016 Article en page(s) : pp 350 - 363 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Termes IGN] Aves
[Termes IGN] biodiversité végétale
[Termes IGN] classification non dirigée
[Termes IGN] couvert forestier
[Termes IGN] données lidar
[Termes IGN] données localisées 3D
[Termes IGN] exploration de données géographiques
[Termes IGN] gestion de la vie sauvage
[Termes IGN] gestion forestière durable
[Termes IGN] habitat d'espèce
[Termes IGN] hauteur des arbres
[Termes IGN] inventaire forestier étranger (données)
[Vedettes matières IGN] Ecologie forestièreRésumé : (auteur) To account for ecological objectives in forest management planning, potential habitats need to be mapped, characterized and evaluated for utility in alternative management practices. Airborne laser scanning (ALS) is increasingly used to derive predictive maps of habitat quality. Unlike ecologically driven approaches that require spatially and temporally co-located training data of the specific species, we tested whether indicative information on the habitat potential could be obtained by means of an unsupervised classification of ALS data. Based on a literature review, altogether five ALS features quantifying vegetation height, cover and diversity were expected to capture the essential variation in the habitat requirements of western capercaillie (Tetrao urogallus L.) and hazel grouse (Tetrastes bonasia L.), which are the most important game birds occurring in the studied area. The features were extracted from sparse density, leaf-off ALS data at a resolution of 256 m2 and partitioned using an unsupervised k-means algorithm. By analysing the persistence of the cluster ensemble formed by the partitioning, altogether 158 plots in 16 structural classes were assigned for field measurements to determine which real-world forest phenomena affected the clustering. The clustering was found to stratify the area mainly in terms of size-related attributes such as timber volume and basal area. The understorey, shrub and herb layers had less correspondence with the clustering, indicating that an unsupervised classification is not directly suitable for habitat mapping. The result was improved using empirical threshold values for the ALS features determined according to the plots labelled as the most potential habitats in the field measurements. This semi-supervised classification of the data indicated 4 per cent of the total forest area as suitable for the specific species, which appears a reasonable estimate of the core area of the habitats considered. Overall, the partitioning formed aggregated, stand-like spatial patterns, even though the neighbourhoods of the individual 256 m2 cells were not considered at all. The result could be further refined by spatial optimization to produce indicative maps for forest management planning with ALS as the sole data source. Numéro de notice : A2016--155 Affiliation des auteurs : non IGN Thématique : FORET Nature : Article DOI : 10.1093/forestry/cpw011 En ligne : https://doi.org/10.1093/forestry/cpw011 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=85780
in Forestry, an international journal of forest research > vol 89 n° 4 (August 2016) . - pp 350 - 363[article]Automatic delineation of built-up area at urban block level from topographic maps / Sebastian Muhs in Computers, Environment and Urban Systems, vol 58 (July 2016)
[article]
Titre : Automatic delineation of built-up area at urban block level from topographic maps Type de document : Article/Communication Auteurs : Sebastian Muhs, Auteur ; Hendrik Herold, Auteur ; Gotthard Meinel, Auteur ; Dirk Burghardt, Auteur ; Odette Kretschmer, Auteur Année de publication : 2016 Article en page(s) : pp 71 - 84 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image
[Termes IGN] analyse d'image numérique
[Termes IGN] base de données historiques
[Termes IGN] carte topographique
[Termes IGN] détection du bâti
[Termes IGN] extraction semi-automatique
[Termes IGN] îlot urbain
[Termes IGN] vision par ordinateur
[Termes IGN] zone urbaineRésumé : (auteur) To comprehensively study and better understand urban dynamic processes — such as densification, growth and sprawl, or shrinkage — spatio-temporal databases that allow to track changes of geographic objects like buildings and urban blocks are essential. While comprehensive databases exist for contemporary data, they usually lack a historic dimension. The manual constitution of historic geographic data, be it based on historic maps or aerial images, is a time consuming and laborious process, however. Therefore, we present an approach to semi-automatically extract this data from binary topographic maps with regard to built-up areas at urban block level. The suitability of topographic maps for historic urban analysis has been proven in previous research. To overcome the challenges that are inherent in scanned topographic maps in regard to digital image interpretation we designed a modular process. Among others, these challenges include fused and (multi-)fragmented map objects caused by the overlap of competing content layers in one single binary map. After a preliminary separation of individual map object layers from the map content, the process follows a two-stage top-down approach. At first, the map is organized into street blocks, which after that are re-delineated in regard to built-up area. In doing so, we achieve correctness values ranging from 0.97 to 0.93 for three study sites in Germany. With an increasing number of projects that provide historic topographic maps as georeferenced digital data, our process represents a promising approach to efficiently prepare these historic data for integration into a spatio-temporal database with minimal user intervention. Numéro de notice : A2016-404 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.compenvurbsys.2016.04.001 En ligne : http://dx.doi.org/10.1016/j.compenvurbsys.2016.04.001 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=81221
in Computers, Environment and Urban Systems > vol 58 (July 2016) . - pp 71 - 84[article]Efficient multiple-feature learning-based hyperspectral image classification with limited training samples / Chongyue Zhao in IEEE Transactions on geoscience and remote sensing, vol 54 n° 7 (July 2016)PermalinkEnabling maps/location searches on mobile devices: constructing a POI database via focused crawling and information extraction / Hsiu-Min Chuang in International journal of geographical information science IJGIS, vol 30 n° 7- 8 (July - August 2016)PermalinkA hierarchical approach to three-dimensional segmentation of LiDAR data at single-tree level in a multilayered forest / Claudia Paris in IEEE Transactions on geoscience and remote sensing, vol 54 n° 7 (July 2016)PermalinkLearning-based superresolution land cover mapping / Feng Ling in IEEE Transactions on geoscience and remote sensing, vol 54 n° 7 (July 2016)PermalinkModeling of ionosphere time series using wavelet neural networks (case study: N-W of Iran) / Mir Reza Ghaffari Razin in Advances in space research, vol 58 n° 1 (July 2016)PermalinkPrediction of categorical spatial data via Bayesian updating / Xiang Huang in International journal of geographical information science IJGIS, vol 30 n° 7- 8 (July - August 2016)PermalinkSparse and low-rank graph for discriminant analysis of hyperspectral imagery / Wei Li in IEEE Transactions on geoscience and remote sensing, vol 54 n° 7 (July 2016)PermalinkThe story of DB4GeO – A service-based geo-database architecture to support multi-dimensional data analysis and visualization / Martin Breunig in ISPRS Journal of photogrammetry and remote sensing, vol 117 (July 2016)PermalinkUsing seal trajectories in biological early warning system for real-time zone tracking / Rouaa Wannous in Ingénierie des systèmes d'information, ISI : Revue des sciences et technologies de l'information, RSTI, vol 21 n° 4 (juillet - août 2016)PermalinkAn assessment of algorithmic parameters affecting image classification accuracy by random forests / Dee Shi in Photogrammetric Engineering & Remote Sensing, PERS, vol 82 n° 6 (June 2016)PermalinkAn intelligent geospatial processing unit for image classification based on geographic vector agents (GVAs) / Kambiz Borna in Transactions in GIS, vol 20 n° 3 (June 2016)PermalinkGrid pattern recognition in road networks using the C4.5 algorithm / Jing Tian in Cartography and Geographic Information Science, Vol 43 n° 3 (June 2016)PermalinkA multilevel point-cluster-based discriminative feature for ALS point cloud classification / Zhenxin Zhang in IEEE Transactions on geoscience and remote sensing, vol 54 n° 6 (June 2016)PermalinkSpatial discovery and the research library / Sara Lafia in Transactions in GIS, vol 20 n° 3 (June 2016)PermalinkAnalysis of human mobility patterns from GPS trajectories and contextual information / Katarzyna Siła-Nowicka in International journal of geographical information science IJGIS, vol 30 n° 5-6 (May - June 2016)PermalinkATLAS: A three-layered approach to facade parsing / Markus Mathias in International journal of computer vision, vol 118 n° 1 (May 2016)PermalinkAutonomous ortho-rectification of very high resolution imagery using SIFT and genetic algorithm / Pramod Kumar Konugurthi in Photogrammetric Engineering & Remote Sensing, PERS, vol 82 n° 5 (May 2016)PermalinkDeep filter banks for texture recognition, description, and segmentation / Mircea Cimpoi in International journal of computer vision, vol 118 n° 1 (May 2016)PermalinkExploiting joint sparsity for pansharpening : the J-SparseFI algorithm / Xiao Xiang Zhu in IEEE Transactions on geoscience and remote sensing, vol 54 n° 5 (May 2016)PermalinkKernel-based domain-invariant feature selection in hyperspectral images for transfer learning / Claudio Persello in IEEE Transactions on geoscience and remote sensing, vol 54 n° 5 (May 2016)Permalink