Random Forests for land cover classification

Gislason, Pall Oskar; Benediktsson, Jón Atli; Sveinsson, Jóhannes R.

doi:10.1016/j.patrec.2005.08.011

Cited by 1,767 publications

(1,027 citation statements)

References 11 publications

Supporting

Mentioning

974

Contrasting

Unclassified

Order By: Relevance

“…RF was selected for this study because it generally outperforms conventional classifiers such as the Gaussian maximum likelihood classifier [61,62], while performing favorably, or equally well, to other non-parametric approaches; e.g., CART [63,64], Support Vector Machines [32,65,66], Artificial Neural Networks [67], and K-Nearest Neighbor [68]. It is a powerful non-linear and non-parametric classifier that allows for fusion and aggregation of high-dimensional data from various sources (e.g., optical, SAR, and topography [30,69,70]; SAR and topography [21,58,71]; and optical and topography [72][73][74]).…”

Section: Image Classificationmentioning

confidence: 99%

Mapping the Dabus Wetlands, Ethiopia, Using Random Forest Classification of Landsat, PALSAR and Topographic Data

et al. 2017

View full text Add to dashboard Cite

Abstract:The Dabus Wetland complex in the highlands of Ethiopia is within the headwaters of the Nile Basin and is home to significant ecological communities and rare or endangered species. Its many interrelated wetland types undergo seasonal and longer-term changes due to weather and climate variations as well as anthropogenic land use such as grazing and burning. Mapping and monitoring of these wetlands has not been previously undertaken due primarily to their relative isolation and lack of resources. This study investigated the potential of remote sensing based classification for mapping the primary vegetation groups in the Dabus Wetlands using a combination of dry and wet season data, including optical (Landsat spectral bands and derived vegetation and wetness indices), radar (ALOS PALSAR L-band backscatter), and elevation (SRTM derived DEM and other terrain metrics) as inputs to the non-parametric Random Forest (RF) classifier. Eight wetland types and three terrestrial/upland classes were mapped using field samples of observed plant community composition and structure groupings as reference information. Various tests to compare results using different RF input parameters and data types were conducted. A combination of multispectral optical, radar and topographic variables provided the best overall classification accuracy, 94.4% and 92.9% for the dry and wet season, respectively. Spectral and topographic data (radar data excluded) performed nearly as well, while accuracies using only radar and topographic data were 82-89%. Relatively homogeneous classes such as Papyrus Swamps, Forested Wetland, and Wet Meadow yielded the highest accuracies while spatially complex classes such as Emergent Marsh were more difficult to accurately classify. The methods and results presented in this paper can serve as a basis for development of long-term mapping and monitoring of these and other non-forested wetlands in Ethiopia and other similar environmental settings.

show abstract

Section: Image Classificationmentioning

confidence: 99%

Mapping the Dabus Wetlands, Ethiopia, Using Random Forest Classification of Landsat, PALSAR and Topographic Data

et al. 2017

View full text Add to dashboard Cite

show abstract

“…The Scikit-learn Python library [28], which requires data to be presented as NumPy arrays, contains a number of machine learning algorithms, including random forests, which has demonstrated good performance when applied to remote sensing data (e.g., [29]). The Scikit-learn implementation of random forests, as with many of the algorithms available, is able to utilize multiple cores for improved performance.…”

Section: Supervised Classificationmentioning

confidence: 99%

A Python-Based Open Source System for Geographic Object-Based Image Analysis (GEOBIA) Utilizing Raster Attribute Tables

et al. 2014

View full text Add to dashboard Cite

A modular system for performing Geographic Object-Based Image Analysis (GEOBIA), using entirely open source (General Public License compatible) software, is presented based around representing objects as raster clumps and storing attributes as a raster attribute table (RAT). The system utilizes a number of libraries, developed by the authors: The Remote Sensing and GIS Library (RSGISLib), the Raster I/O Simplification (RIOS) Python Library, the KEA image format and TuiView image viewer. All libraries are accessed through Python, providing a common interface on which to build processing chains. Three examples are presented, to demonstrate Remote Sens. 2014, 6 6112 the capabilities of the system: (1) classification of mangrove extent and change in French Guiana; (2) a generic scheme for the classification of the UN-FAO land cover classification system (LCCS) and their subsequent translation to habitat categories; and (3) a national-scale segmentation for Australia.The system presented provides similar functionality to existing GEOBIA packages, but is more flexible, due to its modular environment, capable of handling complex classification processes and applying them to larger datasets.

show abstract

“…Random forests is said to be as accurate as or better than adaptive boosting, yet computationally faster (Breiman, 2001;Gislason et al, 2006). Instead of growing just one tree, many (hundreds to thousands) unpruned, independent trees are grown.…”

Section: Soil Spatial Prediction Functionsmentioning

confidence: 99%

Random Forests Applied as a Soil Spatial Predictive Model in Arid Utah

Stum

2010

Digital Soil Mapping

View full text Add to dashboard Cite

Initial soil surveys are incomplete for large tracts of public land in the western USA. Digital soil mapping offers a quantitative approach as an alternative to traditional soil mapping. I sought to predict soil classes across an arid to semiarid watershed of western Utah by applying random forests (RF) and using environmental covariates derived from Landsat 7 Enhanced Thematic Mapper Plus (ETM+) and digital elevation models (DEM). Random forests are similar to classification and regression trees (CART).However, RF is doubly random. Many (e.g., 500) weak trees are grown (trained) independently because each tree is trained with a new randomly selected bootstrap sample, and a random subset of variables is used to split each node. To train and validate the RF trees, 561 soil descriptions were made in the field. An additional 111 points were added by case-based reasoning using aerial photo interpretation. As RF makes classification decisions from the mode of many independently grown trees, model iii uncertainty can be derived. The overall out of the bag (OOB) error was lower without weighting of classes; weighting increased the overall OOB error and the resulting output did not reflect soil-landscape relationships observed in the field. The final RF model had an OOB error of 55.2% and predicted soils on landforms consistent with soil-landscape relationships. The OOB error for individual classes typically decreased with increasing class size. In addition to the final classification, I determined the second and third most likely classification, model confidence, and the hypothetical extent of individual classes.Pixels that had high possibility of belonging to multiple soil classes were aggregated using a minimum confidence value based on limiting soil features, which is an effective and objective method of determining membership in soil map unit associations and complexes mapped at the 1:24,000 scale. Variables derived from both DEM and Landsat 7 ETM+ sources were important for predicting soil classes based on Gini and standard measures of variable importance and OOB errors from groves grown with exclusively DEM-or Landsat-derived data. Random forests was a powerful predictor of soil classes and produced outputs that facilitated further understanding of soil-landscape relationships.(147 pages) iv ACKNOWLEDGMENTS

show abstract

Random Forests for land cover classification

Cited by 1,767 publications

References 11 publications

Mapping the Dabus Wetlands, Ethiopia, Using Random Forest Classification of Landsat, PALSAR and Topographic Data

Mapping the Dabus Wetlands, Ethiopia, Using Random Forest Classification of Landsat, PALSAR and Topographic Data

A Python-Based Open Source System for Geographic Object-Based Image Analysis (GEOBIA) Utilizing Raster Attribute Tables

Random Forests Applied as a Soil Spatial Predictive Model in Arid Utah

Contact Info

Product

Resources

About