Descripteur
Termes IGN > mathématiques > statistique mathématique > analyse de données > classification > classification bayesienne
classification bayesienneSynonyme(s)classification probabiliste bayésienneVoir aussi |
Documents disponibles dans cette catégorie (71)
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
Estimating forest species abundance through linear unmixing of CHRIS/PROBA imagery / S. Stagakis in ISPRS Journal of photogrammetry and remote sensing, vol 119 (September 2016)
[article]
Titre : Estimating forest species abundance through linear unmixing of CHRIS/PROBA imagery Type de document : Article/Communication Auteurs : S. Stagakis, Auteur ; Theofilos Vanikiotis, Auteur ; Olga Sykioti, Auteur Année de publication : 2016 Article en page(s) : pp 79 - 89 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Termes IGN] analyse des mélanges spectraux
[Termes IGN] carte de la végétation
[Termes IGN] classification bayesienne
[Termes IGN] effet d'ombre
[Termes IGN] espèce végétale
[Termes IGN] Fagus sylvatica
[Termes IGN] Grèce
[Termes IGN] image hyperspectrale
[Termes IGN] image Landsat-8
[Termes IGN] image PROBA-CHRIS
[Termes IGN] orthoimage
[Termes IGN] parc naturel national
[Termes IGN] partition d'image
[Termes IGN] Pinus nigra
[Termes IGN] richesse floristique
[Vedettes matières IGN] Inventaire forestierRésumé : (Auteur) The advancing technology of hyperspectral remote sensing offers the opportunity of accurate land cover characterization of complex natural environments. In this study, a linear spectral unmixing algorithm that incorporates a novel hierarchical Bayesian approach (BI-ICE) was applied on two spatially and temporally adjacent CHRIS/PROBA images over a forest in North Pindos National Park (Epirus, Greece). The scope is to investigate the potential of this algorithm to discriminate two different forest species (i.e. beech – Fagus sylvatica, pine – Pinus nigra) and produce accurate species-specific abundance maps. The unmixing results were evaluated in uniformly distributed plots across the test site using measured fractions of each species derived by very high resolution aerial orthophotos. Landsat-8 images were also used to produce a conventional discrete-type classification map of the test site. This map was used to define the exact borders of the test site and compare the thematic information of the two mapping approaches (discrete vs abundance mapping). The required ground truth information, regarding training and validation of the applied mapping methodologies, was collected during a field campaign across the study site. Abundance estimates reached very good overall accuracy (R2 = 0.98, RMSE = 0.06). The most significant source of error in our results was due to the shadowing effects that were very intense in some areas of the test site due to the low solar elevation during CHRIS acquisitions. It is also demonstrated that the two mapping approaches are in accordance across pure and dense forest areas, but the conventional classification map fails to describe the natural spatial gradients of each species and the actual species mixture across the test site. Overall, the BI-ICE algorithm presented increased potential to unmix challenging objects with high spectral similarity, such as different vegetation species, under real and not optimum acquisition conditions. Its full potential remains to be investigated in further and more complex study sites in view of the upcoming satellite hyperspectral missions. Numéro de notice : A2016-778 Affiliation des auteurs : non IGN Thématique : FORET/IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.isprsjprs.2016.05.013 En ligne : https://doi.org/10.1016/j.isprsjprs.2016.05.013 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=82473
in ISPRS Journal of photogrammetry and remote sensing > vol 119 (September 2016) . - pp 79 - 89[article]Local-scale flood mapping on vegetated floodplains from radiometrically calibrated airborne LiDAR data / Radosław Malinowski in ISPRS Journal of photogrammetry and remote sensing, vol 119 (September 2016)
[article]
Titre : Local-scale flood mapping on vegetated floodplains from radiometrically calibrated airborne LiDAR data Type de document : Article/Communication Auteurs : Radosław Malinowski, Auteur ; Bernhard Höfle, Auteur ; Kristina Koenig, Auteur ; et al., Auteur Année de publication : 2016 Article en page(s) : pp 267 - 279 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Lasergrammétrie
[Termes IGN] capteur aérien
[Termes IGN] cartographie des risques
[Termes IGN] classification
[Termes IGN] classification bayesienne
[Termes IGN] classification par arbre de décision
[Termes IGN] coefficient de rétrodiffusion
[Termes IGN] couvert végétal
[Termes IGN] données lidar
[Termes IGN] données localisées 3D
[Termes IGN] étalonnage radiométrique
[Termes IGN] forme d'onde pleine
[Termes IGN] inondation
[Termes IGN] lidar à retour d'onde complète
[Termes IGN] lit majeurRésumé : (Auteur) Knowledge about the magnitude of localised flooding of riverine areas is crucial for appropriate land management and administration at regional and local levels. However, detection and delineation of localised flooding with remote sensing techniques are often hampered on floodplains by the presence of herbaceous vegetation. To address this problem, this study presents the application of full-waveform airborne laser scanning (ALS) data for detection of floodwater extent. In general, water surfaces are characterised by low values of backscattered energy due to water absorption of the infrared laser shots, but the exact strength of the recorded laser pulse depends on the area covered by the targets located within a laser pulse footprint area. To account for this we analysed the physical quantity of radiometrically calibrated ALS data, the backscattering coefficient, in relation to water and vegetation coverage within a single laser footprint. The results showed that the backscatter was negatively correlated to water coverage, and that of the three distinguished classes of water coverage (low, medium, and high) only the class with the largest extent of water cover (>70%) had relatively distinct characteristics that can be used for classification of water surfaces. Following the laser footprint analysis, three classifiers, namely AdaBoost with Decision Tree, Naïve Bayes and Random Forest, were utilised to classify laser points into flooded and non-flooded classes and to derive the map of flooding extent. The performance of the classifiers is highly dependent on the set of laser points features used. Best performance was achieved by combining radiometric and geometric laser point features. The accuracy of flooding maps based solely on radiometric features resulted in overall accuracies of up to 70% and was limited due to the overlap of the backscattering coefficient values between water and other land cover classes. Our point-based classification methods assure a high mapping accuracy (∼89%) and demonstrate the potential of using full-waveform ALS data to detect water surfaces on floodplain areas with limited water surface exposition through the vegetation canopy. Numéro de notice : A2016-785 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.isprsjprs.2016.06.009 En ligne : https://doi.org/10.1016/j.isprsjprs.2016.06.009 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=82499
in ISPRS Journal of photogrammetry and remote sensing > vol 119 (September 2016) . - pp 267 - 279[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)
[article]
Titre : Efficient multiple-feature learning-based hyperspectral image classification with limited training samples Type de document : Article/Communication Auteurs : Chongyue Zhao, Auteur ; Xinbo Gao, Auteur ; Ying Wang, Auteur ; Jie Li, Auteur Année de publication : 2016 Article en page(s) : pp 4052 - 4062 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique
[Termes IGN] apprentissage dirigé
[Termes IGN] classification
[Termes IGN] classification bayesienne
[Termes IGN] extraction
[Termes IGN] image hyperspectrale
[Termes IGN] méthode fondée sur le noyauRésumé : (Auteur) Linearly derived features have been widely used in hyperspectral image classification to find linear separability of certain classes in recent years. Moreover, nonlinearly transformed features are more effective for class discrimination in real analysis scenarios. However, few efforts have attempted to combine both linear and nonlinear features in the same framework even if they can demonstrate some complementary properties. Moreover, conventional multiple-feature learning-based approaches deal with different features equally, which is not reasonable. This paper proposes an efficient multiple-feature learning-based model with adaptive weights for effectively classifying complex hyperspectral images with limited training samples. A new diversity kernel function is proposed first to simulate the vision perception and analysis procedure of human beings. It could simultaneously evaluate the contrast differences of global features and spatial coherence. Since existing multiple-kernel feature models are always time-consuming, we then design a new adaptive weighted multiple kernel learning method. It employs kernel projection, which could lower the dimensionalities and also learn kernel weights to further discriminate the classification boundaries. For combining both linear and nonlinear features, this paper also proposes a novel decision fusion strategy. The method combines linear and multiple kernel features to balance the classification results of different classifiers. The proposed scheme is tested on several hyperspectral data sets and extended to multisource feature classification environment. The experimental results show that the proposed classification method outperforms most of the existing ones and significantly reduces the computational complexity. Numéro de notice : A2016-878 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2016.2535538 En ligne : http://dx.doi.org/10.1109/TGRS.2016.2535538 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=83041
in IEEE Transactions on geoscience and remote sensing > vol 54 n° 7 (July 2016) . - pp 4052 - 4062[article]Geographically weighted evidence combination approaches for combining discordant and inconsistent volunteered geographical information / Alexis Comber in Geoinformatica, vol 20 n° 3 (July - September 2016)
[article]
Titre : Geographically weighted evidence combination approaches for combining discordant and inconsistent volunteered geographical information Type de document : Article/Communication Auteurs : Alexis Comber, Auteur ; Cidália Costa Fonte, Auteur ; Giles M. Foody, Auteur ; Ana-Maria Olteanu-Raimond , Auteur ; et al., Auteur Année de publication : 2016 Article en page(s) : pp 503 – 527 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Géomatique web
[Termes IGN] analyse combinatoire (maths)
[Termes IGN] classification bayesienne
[Termes IGN] classification de Dempster-Shafer
[Termes IGN] classification floue
[Termes IGN] données localisées des bénévoles
[Termes IGN] données multisources
[Termes IGN] incertitude des données
[Termes IGN] occupation du sol
[Termes IGN] pondération
[Termes IGN] qualité des données
[Termes IGN] régression géographiquement pondérée
[Termes IGN] WikimapiaRésumé : (auteur) There is much interest in being able to combine crowdsourced data. One of the critical issues in information sciences is how to combine data or information that are discordant or inconsistent in some way. Many previous approaches have taken a majority rules approach under the assumption that most people are correct most of the time. This paper analyses crowdsourced land cover data generated by the Geo-Wiki initiative in order to infer the land cover present at locations on a 50 km grid. It compares four evidence combination approaches (Dempster-Shafer, Bayes, Fuzzy Sets and Possibility) applied under a geographically weighted kernel with the geographically weighted average approach applied in many current Geo-Wiki analyses. A geographically weighted approach uses a moving kernel under which local analyses are undertaken. The contribution (or salience) of each data point to the analysis is weighted by its distance to the kernel centre, reflecting Tobler’s 1st law of geography. A series of analyses were undertaken using different kernel sizes (or bandwidths). Each of the geographically weighted evidence combination methods generated spatially distributed measures of belief in hypotheses associated with the presence of individual land cover classes at each location on the grid. These were compared with GlobCover, a global land cover product. The results from the geographically weighted average approach in general had higher correspondence with the reference data and this increased with bandwidth. However, for some classes other evidence combination approaches had higher correspondences possibly because of greater ambiguity over class conceptualisations and / or lower densities of crowdsourced data. The outputs also allowed the beliefs in each class to be mapped. The differences in the soft and the crisp maps are clearly associated with the logics of each evidence combination approach and of course the different questions that they ask of the data. The results show that discordant data can be combined (rather than being removed from analysis) and that data integrated in this way can be parameterised by different measures of belief uncertainty. The discussion highlights a number of critical areas for future research. Numéro de notice : A2016-379 Affiliation des auteurs : LASTIG COGIT+Ext (2012-2019) Thématique : GEOMATIQUE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1007/s10707-016-0248-z Date de publication en ligne : 27/02/2016 En ligne : http://dx.doi.org/ 10.1007/s10707-016-0248-z Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=81146
in Geoinformatica > vol 20 n° 3 (July - September 2016) . - pp 503 – 527[article]A joint Gaussian process model for active visual recognition with expertise estimation in crowdsourcing / Chengjiang Long in International journal of computer vision, vol 116 n° 2 (15th January 2016)
[article]
Titre : A joint Gaussian process model for active visual recognition with expertise estimation in crowdsourcing Type de document : Article/Communication Auteurs : Chengjiang Long, Auteur ; Gang Hua, Auteur ; Ashish Kapoor, Auteur Année de publication : 2016 Article en page(s) : pp 136 - 160 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique
[Termes IGN] apprentissage automatique
[Termes IGN] classification bayesienne
[Termes IGN] classification dirigée
[Termes IGN] distribution de Gauss
[Termes IGN] inférence
[Termes IGN] production participative
[Termes IGN] reconnaissance d'objetsRésumé : (auteur) We present a noise resilient probabilistic model for active learning of a Gaussian process classifier from crowds, i.e., a set of noisy labelers. It explicitly models both the overall label noise and the expertise level of each individual labeler with two levels of flip models. Expectation propagation is adopted for efficient approximate Bayesian inference of our probabilistic model for classification, based on which, a generalized EM algorithm is derived to estimate both the global label noise and the expertise of each individual labeler. The probabilistic nature of our model immediately allows the adoption of the prediction entropy for active selection of data samples to be labeled, and active selection of high quality labelers based on their estimated expertise to label the data. We apply the proposed model for four visual recognition tasks, i.e., object category recognition, multi-modal activity recognition, gender recognition, and fine-grained classification, on four datasets with real crowd-sourced labels from the Amazon Mechanical Turk. The experiments clearly demonstrate the efficacy of the proposed model. In addition, we extend the proposed model with the Predictive Active Set Selection Method to speed up the active learning system, whose efficacy is verified by conducting experiments on the first three datasets. The results show our extended model can not only preserve a higher accuracy, but also achieve a higher efficiency. Numéro de notice : A2016--137 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.1007%2Fs11263-015-0834-9 En ligne : https://doi.org/10.1007/s11263-015-0834-9 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=85903
in International journal of computer vision > vol 116 n° 2 (15th January 2016) . - pp 136 - 160[article]Fusion of hyperspectral images and digital surface models for urban object extraction / Janja Avbelj (2016)PermalinkPermalinkActive learning with gaussian process classifier for hyperspectral image classification / Shujing Sun in IEEE Transactions on geoscience and remote sensing, vol 53 n° 4 (April 2015)PermalinkA robust image matching method based on optimized BaySAC / Zhizhong Kang in Photogrammetric Engineering & Remote Sensing, PERS, vol 80 n° 11 (November 2014)PermalinkPermalinkPermalinkRobust position information system for e-navigation : Bayesian pool of errors / Thomas Buret in Navigation aérienne, maritime, spatiale, terrestre, vol 62 n° 243 (janvier 2014)PermalinkHierarchical method of urban building extraction inspired by human perception / Chao Tao in Photogrammetric Engineering & Remote Sensing, PERS, vol 79 n° 12 (December 2013)PermalinkSemisupervised learning of hyperspectral data with unknown land-cover classes / G. Jun in IEEE Transactions on geoscience and remote sensing, vol 51 n° 1 Tome 1 (January 2013)PermalinkA framework for automatic and unsupervised detection of multiple changes in multitemporal images / Francesca Bovolo in IEEE Transactions on geoscience and remote sensing, vol 50 n° 6 (June 2012)Permalink