Descripteur
Termes IGN > mathématiques > statistique mathématique > analyse de données > classification > classification par arbre de décision > classification par forêts d'arbres décisionnels
classification par forêts d'arbres décisionnelsSynonyme(s)Random forests classification forêts aléatoiresVoir aussi |
Documents disponibles dans cette catégorie (303)
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
Exploring the influencing factors in identifying soil texture classes using multitemporal Landsat-8 and Sentinel-2 data / Yanan Zhou in Remote sensing, vol 14 n° 21 (November-1 2022)
[article]
Titre : Exploring the influencing factors in identifying soil texture classes using multitemporal Landsat-8 and Sentinel-2 data Type de document : Article/Communication Auteurs : Yanan Zhou, Auteur ; Wei Wu, Auteur ; Hongbin Liu, Auteur Année de publication : 2022 Article en page(s) : n° 5571 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique
[Termes IGN] carte d'occupation du sol
[Termes IGN] classification par forêts d'arbres décisionnels
[Termes IGN] classification par séparateurs à vaste marge
[Termes IGN] composition des sols
[Termes IGN] données multitemporelles
[Termes IGN] Extreme Gradient Machine
[Termes IGN] Fleuve bleu (Chine)
[Termes IGN] image Landsat-8
[Termes IGN] image Sentinel-MSI
[Termes IGN] limon
[Termes IGN] qualité du sol
[Termes IGN] réflectance spectrale
[Termes IGN] texture du solRésumé : (auteur) Soil texture is a key soil property driving physical, chemical, biological, and hydrological processes in soils. The rapid development of remote sensing techniques shows great potential for mapping soil properties. This study highlights the effectiveness of multitemporal remote sensing data in identifying soil textural class by using retrieved vegetation properties as proxies of soil properties. The impacts of sensors, modeling resolutions, and modeling techniques on the accuracy of soil texture classification were explored. Multitemporal Landsat-8 and Sentinel-2 images were individually acquired at the same time periods. Three satellite-based experiments with different inputs, i.e., Landsat-8 data, Sentinel-2 data (excluding red-edge parameters), and Sentinel-2 data (including red-edge parameters) were conducted. Modeling was carried out at three spatial resolutions (10, 30, 60 m) using five machine-learning (ML) methods: random forest, support vector machine, gradient-boosting decision tree, categorical boosting, and super learner that combined the four former classifiers based on the stacking concept. In addition, a novel SHapley Addictive Explanation (SHAP) technique was introduced to explain the outputs of the ML model. The results showed that the sensors, modeling resolutions, and modeling techniques significantly affected the prediction accuracy. The models using Sentinel-2 data with red-edge parameters performed consistently best. The models usually gave better results at fine (10 m) and medium (30 m) modeling resolutions than at a coarse (60 m) resolution. The super learner provided higher accuracies than other modeling techniques and gave the highest values of overall accuracy (0.8429), kappa (0.7611), precision (0.8378), recall rate (0.8393), and F1-score (0.8398) at 30 m with Sentinel-2 data involving red-edge parameters. The SHAP technique quantified the contribution of each variable for different soil textural classes, revealing the critical roles of red-edge parameters in separating loamy soils. This study provides comprehensive insights into the effective modeling of soil properties on various scales using multitemporal optical images. Numéro de notice : A2022-856 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.3390/rs14215571 Date de publication en ligne : 04/11/2022 En ligne : https://doi.org/10.3390/rs14215571 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=102104
in Remote sensing > vol 14 n° 21 (November-1 2022) . - n° 5571[article]Features predisposing forest to bark beetle outbreaks and their dynamics during drought / M. Müller in Forest ecology and management, vol 523 (November-1 2022)
[article]
Titre : Features predisposing forest to bark beetle outbreaks and their dynamics during drought Type de document : Article/Communication Auteurs : M. Müller, Auteur ; P.O. Olsson, Auteur ; Lars Eklundh, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : n° 120480 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Termes IGN] analyse des risques
[Termes IGN] canopée
[Termes IGN] caractérisation
[Termes IGN] changement climatique
[Termes IGN] classification par forêts d'arbres décisionnels
[Termes IGN] données météorologiques
[Termes IGN] humidité du sol
[Termes IGN] peuplement mélangé
[Termes IGN] Picea abies
[Termes IGN] Scolytinae
[Termes IGN] sécheresse
[Termes IGN] Suède
[Vedettes matières IGN] Végétation et changement climatiqueRésumé : (auteur) Climate change is estimated to increase the risk of the bark beetle (Ips typographus L.) mass outbreaks in Norway Spruce (Picea abies (L.) Karst) forests. Habitats that are thermally suitable for bark beetles may expand, and an increase in the frequency and intensity of droughts can promote drought stress on host trees. Drought affects tree vigor and in unison with environmental features it influences the local predisposition risk of forest stands to bark beetle attacks. We aimed to study how various environmental features influence the risk of bark beetle attacks during a drought year and the following years with more normal weather conditions but with higher bark beetle populations. We included features representing local forest stand attributes, topography, soil type and wetness, the proximity of clear-cuts and previous bark beetle attacks, and a machine learning algorithm (random forest) was applied to study the variation of predisposition risk across a 48,600 km2 study area in SE Sweden. Forest stands with increased risk of bark beetle attack were distinguished with high accuracy both during drought and in normal weather conditions. The results show that during both study periods, spruce and mixed coniferous forests had elevated risk of attack, while forests with a mix of deciduous and coniferous trees had a lower risk. Forests with high average canopy height were strongly predisposed to bark beetle attacks. However, during the drought year risk was more similar between stands with lower and higher canopy height, suggesting that during drought periods younger trees can be predisposed to bark beetle attacks. The importance of soil moisture and position within the local landscape were highlighted as important features during the drought year. Identifying areas with increased risk, supported by information on how environmental features control the predisposition risk during drought, could aid adaptation strategies and forest management intervention efforts. We conclude that geospatial data and machine learning have the potential to further support the digitalization of the forest industry, facilitating development of methods capable to quantify importance and dynamics of
environmental features controlling the risk in local context. Corresponding methods could help to direct management actions more effectively and offer information for decision-making in changing climate.Numéro de notice : A2022-731 Affiliation des auteurs : non IGN Thématique : FORET Nature : Article DOI : 10.1016/j.foreco.2022.120480 Date de publication en ligne : 07/09/2022 En ligne : https://doi.org/10.1016/j.foreco.2022.120480 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=101687
in Forest ecology and management > vol 523 (November-1 2022) . - n° 120480[article]A deep 2D/3D Feature-Level fusion for classification of UAV multispectral imagery in urban areas / Hossein Pourazar in Geocarto international, vol 37 n° 23 ([15/10/2022])
[article]
Titre : A deep 2D/3D Feature-Level fusion for classification of UAV multispectral imagery in urban areas Type de document : Article/Communication Auteurs : Hossein Pourazar, Auteur ; Farhad Samadzadegan, Auteur ; Farzaneh Dadrass Javan, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : pp 6695 - 6712 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique
[Termes IGN] alignement des données
[Termes IGN] classification par forêts d'arbres décisionnels
[Termes IGN] classification par réseau neuronal convolutif
[Termes IGN] image captée par drone
[Termes IGN] image multibande
[Termes IGN] image proche infrarouge
[Termes IGN] image RVB
[Termes IGN] modèle numérique de surface
[Termes IGN] orthophotoplan numérique
[Termes IGN] zone urbaineRésumé : (auteur) In this paper, a deep convolutional neural network (CNN) is developed to classify the Unmanned Aerial Vehicle (UAV) derived multispectral imagery and normalized digital surface model (DSM) data in urban areas. For this purpose, a multi-input deep CNN (MIDCNN) architecture is designed using 11 parallel CNNs; 10 deep CNNs to extract the features from all possible triple combinations of spectral bands as well as one deep CNN dedicated to the normalized DSM data. The proposed method is compared with the traditional single-input (SI) and double-input (DI) deep CNN designations and random forest (RF) classifier, and evaluated using two independent test datasets. The results indicate that increasing the CNN layers parallelly augmented the classifier’s generalization and reduced overfitting risk. The overall accuracy and kappa value of the proposed method are 95% and 0.93, respectively, for the first test dataset, and 96% and 0.94, respectively, for the second test data set. Numéro de notice : A2022-749 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1080/10106049.2021.1959655 Date de publication en ligne : 04/08/2021 En ligne : https://doi.org/10.1080/10106049.2021.1959655 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=101741
in Geocarto international > vol 37 n° 23 [15/10/2022] . - pp 6695 - 6712[article]Land use/land cover mapping from airborne hyperspectral images with machine learning algorithms and contextual information / Ozlem Akar in Geocarto international, vol 37 n° 22 ([10/10/2022])
[article]
Titre : Land use/land cover mapping from airborne hyperspectral images with machine learning algorithms and contextual information Type de document : Article/Communication Auteurs : Ozlem Akar, Auteur ; Esra Tunc Gormus, Auteur Année de publication : 2022 Article en page(s) : pp 6643 - 6670 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique
[Termes IGN] carte d'occupation du sol
[Termes IGN] carte de la végétation
[Termes IGN] classification orientée objet
[Termes IGN] classification par forêts d'arbres décisionnels
[Termes IGN] classification par séparateurs à vaste marge
[Termes IGN] filtre de Gabor
[Termes IGN] image hyperspectrale
[Termes IGN] matrice de co-occurrence
[Termes IGN] niveau de gris (image)
[Termes IGN] texture d'image
[Termes IGN] transformation en ondelettes
[Termes IGN] TurquieRésumé : (auteur) Land use and Land cover (LULC) mapping is one of the most important application areas of remote sensing which requires both spectral and spatial resolutions in order to decrease the spectral ambiguity of different land cover types. Airborne hyperspectral images are among those data which perfectly suits to that kind of applications because of their high number of spectral bands and the ability to see small details on the field. As this technology has newly developed, most of the image processing methods are for the medium resolution sensors and they are not capable of dealing with high resolution images. Therefore, in this study a new framework is proposed to improve the classification accuracy of land use/cover mapping applications and to achieve a greater reliability in the process of mapping land use map using high resolution hyperspectral image data. In order to achieve it, spatial information is incorporated together with spectral information by exploiting feature extraction methods like Grey Level Co-occurrence Matrix (GLCM), Gabor and Morphological Attribute Profile (MAP) on dimensionally reduced image with highest accuracy. Then, machine learning algorithms like Random Forest (RF) and Support Vector Machine (SVM) are used to investigate the contribution of texture information in the classification of high resolution hyperspectral images. In addition to that, further analysis is conducted with object based RF classification to investigate the contribution of contextual information. Finally, overall accuracy, producer’s/user’s accuracy, the quantity and allocation based disagreements and location and quantity based kappa agreements are calculated together with McNemar tests for the accuracy assessment. According to our results, proposed framework which incorporates Gabor texture information and exploits Discrete Wavelet Transform based dimensionality reduction method increase the overall classification accuracy up to 9%. Amongst individual classes, Gabor features boosted classification accuracies of all the classes (soil, road, vegetation, building and shadow) to 7%, 6%, 6%, 8%, 9%, and 24% respectively with producer’s accuracy. Besides, 17% and 10% increase obtained in user’s accuracy with MAP (area) feature in classifying road and shadow classes respectively. Moreover, when the object based classification is conducted, it is seen that the OA of pixel based classification is increased further by 1.07%. An increase between 2% and 4% is achieved with producer’s accuracy in soil, vegetation and building classes and an increase between 1% and 3% is achieved by user’s accuracy in soil, road, vegetation and shadow classes. In the end, accurate LULC map is produced with object based RF classification of gabor features added airborne hyperspectral image which is dimensionally reduced with DWT method. Numéro de notice : A2022-729 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1080/10106049.2021.1944453 Date de publication en ligne : 09/11/2021 En ligne : https://doi.org/10.1080/10106049.2021.1944453 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=101675
in Geocarto international > vol 37 n° 22 [10/10/2022] . - pp 6643 - 6670[article]Application of a graph convolutional network with visual and semantic features to classify urban scenes / Yongyang Xu in International journal of geographical information science IJGIS, vol 36 n° 10 (October 2022)
[article]
Titre : Application of a graph convolutional network with visual and semantic features to classify urban scenes Type de document : Article/Communication Auteurs : Yongyang Xu, Auteur ; Shuai Jin, Auteur ; Zhanlong Chen, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : pp 2009-2034 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Géomatique
[Termes IGN] classification par forêts d'arbres décisionnels
[Termes IGN] classification par réseau neuronal convolutif
[Termes IGN] classification par séparateurs à vaste marge
[Termes IGN] extraction de traits caractéristiques
[Termes IGN] matrice de co-occurrence
[Termes IGN] OpenStreetMap
[Termes IGN] Pékin (Chine)
[Termes IGN] point d'intérêt
[Termes IGN] relation spatiale
[Termes IGN] réseau neuronal de graphes
[Termes IGN] réseau routier
[Termes IGN] scène urbaineRésumé : (auteur) Urban scenes consist of visual and semantic features and exhibit spatial relationships among land-use types (e.g. industrial areas are far away from the residential zones). This study applied a graph convolutional network with neighborhood information (henceforth, named the neighbour supporting graph convolutional neural network), to learn spatial relationships for urban scene classification. Furthermore, a co-occurrence analysis with visual and semantic features proceeded to improve the accuracy of urban scene classification. We tested the proposed method with the fifth ring road of Beijing with an overall classification accuracy of 0.827 and a Kappa coefficient of 0.769. In comparison with other methods, such as support vector machine, random forest, and general graph convolutional network, the case study showed that the proposed method improved about 10% in urban scene classification. Numéro de notice : A2022-740 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE/URBANISME Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1080/13658816.2022.2048834 Date de publication en ligne : 10/03/2022 En ligne : https://doi.org/10.1080/13658816.2022.2048834 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=101717
in International journal of geographical information science IJGIS > vol 36 n° 10 (October 2022) . - pp 2009-2034[article]Comparison of layer-stacking and Dempster-Shafer theory-based methods using Sentinel-1 and Sentinel-2 data fusion in urban land cover mapping / Dang Hung Bui in Geo-spatial Information Science, vol 25 n° 3 (October 2022)PermalinkDetecting overmature forests with airborne laser scanning (ALS) / Marc Fuhr in Remote sensing in ecology and conservation, vol 8 n° 5 (October 2022)PermalinkA comparative assessment of modeling groundwater vulnerability using DRASTIC method from GIS and a novel classification method using machine learning classifiers / Qasim Khan in Geocarto international, vol 37 n° 20 ([20/09/2022])PermalinkDevelopment of a novel hybrid multi-boosting neural network model for spatial prediction of urban flood / Amid Darabi in Geocarto international, vol 37 n° 19 ([15/09/2022])PermalinkForest canopy stratification based on fused, imbalanced and collinear LiDAR and Sentinel-2 metrics / Jakob Wernicke in Remote sensing of environment, vol 279 (September-15 2022)PermalinkAnalytical method for high-precision seabed surface modelling combining B-spline functions and Fourier series / Tyler Susa in Marine geodesy, vol 45 n° 5 (September 2022)PermalinkForest tree species classification based on Sentinel-2 images and auxiliary data / Haotian You in Forests, vol 13 n° 9 (september 2022)PermalinkIdentification of urban sectors prone to solid waste accumulation: A machine learning approach based on social indicators / Luis Izquierdo-Horna in Computers, Environment and Urban Systems, vol 96 (September 2022)PermalinkMapping individual abandoned houses across cities by integrating VHR remote sensing and street view imagery / Shengyuan Zou in International journal of applied Earth observation and geoinformation, vol 113 (September 2022)PermalinkPoint-of-interest detection from Weibo data for map updating / Xue Yang in Transactions in GIS, vol 26 n° 6 (September 2022)Permalink