Centre de documentation
scientifique

National scale mapping of larch plantations for Wales using the Sentinel-2 data archive / Suvarna M. Punalekar in Forest ecology and management, vol 501 (December-1 2021)  

Public [article]  inForest ecology and management > vol 501 (December-1 2021) . - n° 119679 Titre : National scale mapping of larch plantations for Wales using the Sentinel-2 data archive Type de document : Article/Communication Auteurs : Suvarna M. Punalekar, Auteur ; Carole Planque, Auteur ; Richard M. Lucas, Auteur ; et al., Auteur Année de publication : 2021 Article en page(s) : n° 119679 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection [Termes IGN] apprentissage automatique [Termes IGN] arbre de décision [Termes IGN] carte forestière [Termes IGN] coupe rase (sylviculture) [Termes IGN] gestion forestière [Termes IGN] image infrarouge [Termes IGN] image Sentinel-MSI [Termes IGN] indice de végétation [Termes IGN] Larix decidua [Termes IGN] maladie phytosanitaire [Termes IGN] modélisation de la forêt [Termes IGN] Pays de Galles [Termes IGN] surveillance forestière Résumé : (auteur) Accurate spatial information regarding forest types and tree species is immensely important for efficient forest management strategies. In the UK and particularly in Wales, creating a spatial inventory of larch (Larix sps.) plantations that encompasses both the public and private forests has become one of the highest priorities of woodland management policies, particularly given the need to respond to the rapid spread of Phytophthora ramorum fungal disease. For directing disease control measures, national scale, regularly updated mapping of larch distributions is essential. In this study, we applied a ExtraTree classifier machine learning algorithm to multi-year (June 2015 and December 2019) multi-path composites of vegetation indices derived from 10 m Sentinel-2 satellite data (spectral range used in this study: 490–2190 nm) to map the extent of larch plantations across Wales. For areas identified as woody vegetation, areas under larch plantations were associated with a needle-leaved leaf type and deciduous phenology, allowing differentiation from broad-leaved deciduous and needle-leaved evergreen types. The model accuracies for validation, which included overall accuracy, producer’s and user’s accuracies, exceeded 95% and the F1-score was greater than 0.97 for all forest types. Comparison against an independent reference dataset indicated all map accuracies above 90% (F1-score higher than 0.92) with the lowest value being 90.3% for the producer’s accuracy for larch. Short wave infrared and red-edge based indices were particularly useful for discriminating larch from other forest types. Capacity for updating information on clear-felling of larch stands through annual updates of a woody mask was also introduced. The resulting maps of larch plantations for Wales are the most current for Wales covering public as well as private woodlands and can be routinely updated. The classification approach has potential to be transferred to a wider geographical area given the availability of open-source multi-year Sentienl-2 datasets and robust calibration datasets. Numéro de notice : A2021-741 Affiliation des auteurs : non IGN Thématique : FORET/IMAGERIE Nature : Article DOI : 10.1016/j.foreco.2021.119679 Date de publication en ligne : 20/09/2021 En ligne : https://doi.org/10.1016/j.foreco.2021.119679 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=98657 [article]

Bagging and boosting ensemble classifiers for classification of multispectral, hyperspectral and PolSAR data: A comparative evaluation / Hamid Jafarzadeh in Remote sensing, vol 13 n° 21 (November-1 2021)  

Public [article]  inRemote sensing > vol 13 n° 21 (November-1 2021) . - n° 4405 Titre : Bagging and boosting ensemble classifiers for classification of multispectral, hyperspectral and PolSAR data: A comparative evaluation Type de document : Article/Communication Auteurs : Hamid Jafarzadeh, Auteur ; Masoud Mahdianpari, Auteur ; Eric Gill, Auteur ; et al., Auteur Année de publication : 2021 Article en page(s) : n° 4405 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image mixte [Termes IGN] analyse comparative [Termes IGN] apprentissage automatique [Termes IGN] arbre de décision [Termes IGN] boosting adapté [Termes IGN] classification par forêts d'arbres décisionnels [Termes IGN] données polarimétriques [Termes IGN] ensachage [Termes IGN] Extreme Gradient Machine [Termes IGN] image hyperspectrale [Termes IGN] image multibande [Termes IGN] image radar moirée [Termes IGN] image ROSIS Résumé : (auteur) In recent years, several powerful machine learning (ML) algorithms have been developed for image classification, especially those based on ensemble learning (EL). In particular, Extreme Gradient Boosting (XGBoost) and Light Gradient Boosting Machine (LightGBM) methods have attracted researchers’ attention in data science due to their superior results compared to other commonly used ML algorithms. Despite their popularity within the computer science community, they have not yet been well examined in detail in the field of Earth Observation (EO) for satellite image classification. As such, this study investigates the capability of different EL algorithms, generally known as bagging and boosting algorithms, including Adaptive Boosting (AdaBoost), Gradient Boosting Machine (GBM), XGBoost, LightGBM, and Random Forest (RF), for the classification of Remote Sensing (RS) data. In particular, different classification scenarios were designed to compare the performance of these algorithms on three different types of RS data, namely high-resolution multispectral, hyperspectral, and Polarimetric Synthetic Aperture Radar (PolSAR) data. Moreover, the Decision Tree (DT) single classifier, as a base classifier, is considered to evaluate the classification’s accuracy. The experimental results demonstrated that the RF and XGBoost methods for the multispectral image, the LightGBM and XGBoost methods for hyperspectral data, and the XGBoost and RF algorithms for PolSAR data produced higher classification accuracies compared to other ML techniques. This demonstrates the great capability of the XGBoost method for the classification of different types of RS data. Numéro de notice : A2021-823 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.3390/rs13214405 Date de publication en ligne : 02/11/2021 En ligne : https://doi.org/10.3390/rs13214405 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=98938 [article]

A comparison of a gradient boosting decision tree, random forests, and artificial neural networks to model urban land use changes: the case of the Seoul metropolitan area / Myung-Jin Jun in International journal of geographical information science IJGIS, vol 35 n° 11 (November 2021)  

Public [article]  inInternational journal of geographical information science IJGIS > vol 35 n° 11 (November 2021) . - pp 2149 - 2167 Titre : A comparison of a gradient boosting decision tree, random forests, and artificial neural networks to model urban land use changes: the case of the Seoul metropolitan area Type de document : Article/Communication Auteurs : Myung-Jin Jun, Auteur Année de publication : 2021 Article en page(s) : pp 2149 - 2167 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Analyse spatiale [Termes IGN] analyse comparative [Termes IGN] arbre de décision [Termes IGN] changement d'utilisation du sol [Termes IGN] classification par forêts d'arbres décisionnels [Termes IGN] Extreme Gradient Machine [Termes IGN] modèle de simulation [Termes IGN] réseau neuronal artificiel [Termes IGN] Séoul [Termes IGN] zone urbaine Résumé : (auteur) This study compares the performance of gradient boosting decision tree (GBDT), artificial neural networks (ANNs), and random forests (RF) methods in LUC modeling in the Seoul metropolitan area. The results of this study showed that GBDT and RF have higher predictive power than ANN, indicating that tree-based ensemble methods are an effective technique for LUC prediction. Along with the outstanding predictive performance, the DT-based ensemble models provide insights for understanding which factors drive LUCs in complex urban dynamics with the relative importance and nonlinear marginal effects of predictor variables. The GBDT results indicate that distance to the existing residential site has the highest contribution to urban land use conversion (30.4% of the relative importance), while other significant predictor variables were proximity to industrial and public sites (combined 32.3% of relative importance). New residential development is likely to be adjacent to existing residential sites, but nonresidential development occurs at a distance (about 600 m) from such sites. The distance to the central business district (CBD) had increasing marginal effects on residential land use conversion, while no significant pattern was found for nonresidential land use conversion, indicating that Seoul has experienced more population suburbanization than employment decentralization. Numéro de notice : A2021-756 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1080/13658816.2021.1887490 Date de publication en ligne : 01/03/2021 En ligne : https://doi.org/10.1080/13658816.2021.1887490 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=98771 [article]

Exemplaires(1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
079-2021111	SL	Revue	Centre de documentation	Revues en salle	Disponible

Diffuse attenuation coefficient (Kd) from ICESat-2 ATLAS spaceborne Lidar using random-forest regression / Forrest Corcoran in Photogrammetric Engineering & Remote Sensing, PERS, vol 87 n° 11 (November 2021)  

Public [article]  inPhotogrammetric Engineering & Remote Sensing, PERS > vol 87 n° 11 (November 2021) . - pp 831 - 840 Titre : Diffuse attenuation coefficient (Kd) from ICESat-2 ATLAS spaceborne Lidar using random-forest regression Type de document : Article/Communication Auteurs : Forrest Corcoran, Auteur ; Christopher E. Parrish, Auteur Année de publication : 2021 Article en page(s) : pp 831 - 840 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Lasergrammétrie [Termes IGN] apprentissage automatique [Termes IGN] arbre de décision [Termes IGN] capteur spatial [Termes IGN] classification par forêts d'arbres décisionnels [Termes IGN] données ICEsat [Termes IGN] données lidar [Termes IGN] données localisées 3D [Termes IGN] forme d'onde [Termes IGN] littoral [Termes IGN] modèle de régression [Termes IGN] semis de points [Termes IGN] turbidité des eaux Résumé : (Auteur) This study investigates a new method for measuring water turbidity—specifically, the diffuse attenuation coefficient of downwelling irradiance Kd —using data from a spaceborne, green-wavelength lidar aboard the National Aeronautics and Space Administration's ICESat-2 satellite. The method enables us to fill nearshore data voids in existing Kd data sets and provides a more direct measurement approach than methods based on passive multispectral satellite imagery. Furthermore, in contrast to other lidar-based methods, it does not rely on extensive signal processing or the availability of the system impulse response function, and it is designed to be applied globally rather than at a specific geographic location. The model was tested using Kd measurements from the National Oceanic and Atmospheric Administration's Visible Infrared Imaging Radiometer Suite sensor at 94 coastal sites spanning the globe, with Kd values ranging from 0.05 to 3.6 m –1 . The results demonstrate the efficacy of the approach and serve as a benchmark for future machine-learning regression studies of turbidity using ICESat-2. Numéro de notice : A2021-896 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.14358/PERS.21-00013R2 Date de publication en ligne : 01/11/2021 En ligne : https://doi.org/10.14358/PERS.21-00013R2 Format de la ressource électronique : URL Article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=99272 [article]

Exemplaires(1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
105-2021111	SL	Revue	Centre de documentation	Revues en salle	Disponible

Two hidden layer neural network-based rotation forest ensemble for hyperspectral image classification / Laxmi Narayana Eeti in Geocarto international, vol 36 n° 16 ([01/09/2021])  

Public [article]  inGeocarto international > vol 36 n° 16 [01/09/2021] . - pp 1820 - 1837 Titre : Two hidden layer neural network-based rotation forest ensemble for hyperspectral image classification Type de document : Article/Communication Auteurs : Laxmi Narayana Eeti, Auteur ; Krishna Mohan Buddhiraju, Auteur Année de publication : 2021 Article en page(s) : pp 1820 - 1837 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] arbre de décision [Termes IGN] classification par forêts d'arbres décisionnels [Termes IGN] ensachage [Termes IGN] image AVIRIS [Termes IGN] image EO1-Hyperion [Termes IGN] image hyperspectrale [Termes IGN] image ROSIS [Termes IGN] Perceptron multicouche [Termes IGN] précision de la classification [Termes IGN] réseau neuronal profond [Termes IGN] Rotation Forest classification Résumé : (auteur) Decision tree-based Rotation Forest could generate satisfactory but lower classification accuracy for a given training sample set and image data, owing to the inherent disadvantages in decision trees, namely myopic, replication and fragmentation problem. To improve performance of Rotation Forest technique, we propose to utilize two-hidden-layered-feedforward neural network as base classifier instead of decision tree. We examine the classification performance of proposed model under two situations, namely when free network parameters are maintained the same across all ensemble components and otherwise. The proposed model, where each component is initialized with different pair of initial weights and bias, performs better than decision tree-based Rotation Forest on three different Hyperspectral sensor datasets – AVIRIS, ROSIS and Hyperion. Improvements in classification accuracy are above 2% and up to 3% depending upon dataset. Also, the proposed model achieves improvement in accuracy over Random Forest in the range 4.2–8.8%. Numéro de notice : A2021-581 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1080/10106049.2019.1678680 Date de publication en ligne : 21/10/2019 En ligne : https://doi.org/10.1080/10106049.2019.1678680 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=98193 [article]

An adaptive filtering algorithm of multilevel resolution point cloud / Youyuan Li in Survey review, Vol 53 n° 379 (July 2021)  

Permalink

Machine learning for inference: using gradient boosting decision tree to assess non-linear effects of bus rapid transit on house prices / Linchuan Yang in Annals of GIS, vol 27 n° 3 (July 2021)  

Permalink

Predicting tree species based on the geometry and density of aerial laser scanning point cloud of treetops / Nina Kranjec in Geodetski vestnik, vol 65 n° 2 (June - August 2021)  

Permalink

The delineation of tea gardens from high resolution digital orthoimages using mean-shift and supervised machine learning methods / Akhtar Jamil in Geocarto international, vol 36 n° 7 ([15/04/2021])  

Permalink

Data mining / Derya Birant (2021)  

Permalink

Extraction of street pole-like objects based on plane filtering from mobile LiDAR data / Jingming Tu in IEEE Transactions on geoscience and remote sensing, vol 59 n° 1 (January 2021)  

Permalink

LiDAR-based point clouds registration for localization in indoor environments / Ketty Favre (2021)  

Permalink

A method of hydrographic survey technology selection based on the decision tree supervised learning / Ivana Golub Medvešek (2021)  

Permalink

Forêt d'arbres aléatoires et classification d'images satellites : relation entre la précision du modèle d'entraînement et la précision globale de la classification / Aurélien N.G. Matsaguim in Revue Française de Photogrammétrie et de Télédétection, n° 222 (novembre 2020)  

Permalink

Pairwise coarse registration of point clouds in urban scenes using voxel-based 4-planes congruent sets / Yusheng Xu in ISPRS Journal of photogrammetry and remote sensing, vol 151 (May 2019)  

Permalink