Random Forests for land cover classification

By P. Gislason, Jon Atli Benediktsson, Johannes R. Sveinsson. Pattern Recognition Letters, 2006

2,031 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/7ad785d8cd5d45ee07a82a2c0d7d4c733b1251d5

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

Random forest predictive modeling of mineral prospectivity with small number of prospects and data with missing values in Abra (Philippines) (2015, background): “* Email for correspondence: john.carranza@jcu.edu.au”
Relevance of airborne lidar and multispectral image data for urban scene classification using Random Forests (2011, methodology): “It was successfully applied to multispectral data (Pal, 2005), multitemporal SAR images (Waske and Braun, 2009), hyperspectral data (Ham et al., 2005), or multi-source data (Gislason et al., 2006), where Landsat MSS and topographic data were used.”
A 30-m landsat-derived cropland extent product of Australia and China using random forest machine learning algorithm on Google Earth Engine cloud computing platform (2018, positive): “Pixel-based approaches include: a) Random forest algorithm (Tatsumi et al., 2015; Wang et al., 2015; Gislason et al, 2006); (b) Support vector machines (Mountrakis et al., 2011; Shao and Lunetta, 2012); (c) decision tree algorithms (Ozdogan and Gutman, 2008; Waldner et al., 2016); (d) Tassel cap…”
A Multilevel Multimodal Fusion Transformer for Remote Sensing Semantic Segmentation (2024, methodology): “In the literature, various semantic segmentation methods have been proposed by using random forest [10], support vector machine [11], [12], and conditional random field [13].”
Discrimination of crop types with TerraSAR-X-derived information (2015, methodology): “…widely used 95 supervised learning models – support vector machine (Bovolo et al. 2010; Foody and Mathur 2004; 96 Lizarazo 2008; Pal 2008) and random forest (Duro et al. 2012; Gislason et al. 2006; Kavzoglu and 97 Colkesen 2013; Pal 2005; Rodriguez-Galiano et al. 2012) – were used in this study.”
Land-use/cover classification in a heterogeneous coastal landscape using RapidEye imagery: evaluating the performance of random forest and support vector machines classifiers (2014, methodology): “…using commonly used multi-spectral imagery (Cingolani et al. 2004; Erbek, Özkan, and Taberner 2004; Otukei and Blaschke 2010; Ustin et al. 1996; Gislason, Benediktsson, and Sveinsson 2006; Rodriguez-Galiano et al. 2012; Duro, Franklin, and Dubé 2012; Sanchez-Hernandez, Boyd, and Foody 2007;…”
Estimating smallholder crops production at village level from Sentinel-2 time series in Mali's cotton belt (2018, methodology): “The RF classifier was chosen based on its lack of over-fitting, its parsimony for user-defined parameters, the features importance analysis, its low sensitivity to the number of inputs, its minimization of correlations between classifiers, its light computational requirements, and its robustness to ”
Deep Learning for Change Detection in Remote Sensing Images: Comprehensive Review and Meta-Analysis (2020, methodology): “the analysis of models that either require a greater composition of learned concepts or functions, compared to conventional machine learning models such as naive Bayes [2], [3], support vector machine (SVM) [4], [5], random forests, [6], [7] and the decision tree [8], [9].”
Multisource Remote Sensing Data Classification Based on Convolutional Neural Network (2018, background): “Random forest, with the strategy of boot-bagging [13], [28], provides satisﬁed performance in remote sensing data classiﬁcation owing to its tolerance on big data.”
Deep Learning Classification of Land Cover and Crop Types Using Remote Sensing Data (2017, background): “Another popular approach in the RS domain is the random forest (RF)-based approach [12].”