User Tools

Site Tools


remote_sensing_methods:random_forests

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
remote_sensing_methods:random_forests [2013/10/16 12:59]
jgh [References]
remote_sensing_methods:random_forests [2013/10/16 13:02] (current)
jgh
Line 8: Line 8:
  
 ===== Description ===== ===== Description =====
-Random ​forests ​is a ensemble learning algorithm for regression and classification. ​ Since it is a machine-learning method, random forests is non-parametric because it is not based on any assumptions about data distribution. ​ Unlike statistical approaches, machine-learning is data driven by the relationship between independent and dependent variables (Breiman, 2001). ​  ​Random forests are a combination of tree predictors such that each tree depends on the values of a random vector sampled independently and with the same distribution for all trees in the forest.+Random ​Forests ​is a ensemble learning algorithm for regression and classification. ​ Since it is a machine-learning method, random forests is non-parametric because it is not based on any assumptions about data distribution. ​ Unlike statistical approaches, machine-learning is data driven by the relationship between independent and dependent variables (Breiman, 2001). ​  ​Random forests are a combination of tree predictors such that each tree depends on the values of a random vector sampled independently and with the same distribution for all trees in the forest.
  
 {{:​remote_sensing_methods:​random_forests_1.png}} {{:​remote_sensing_methods:​random_forests_1.png}}
Line 19: Line 19:
  
 ===== Potential Remote Sensing Applications ===== ===== Potential Remote Sensing Applications =====
-The following three cases are a small sample of the many ways the Random ​Forest ​algorithm has been applied to remotely sensed image classification. ​  +The following three cases are a small sample of the many ways the Random ​Forests ​algorithm has been applied to remotely sensed image classification. ​  
  
-In a comparison of random forests ​with three other classification algorithms (Gentle AdaBoost [GAB], Support Vector Machine [SVM] and Maximum Likelihood Classification [MLC]), initial findings indicate that random forests gives higher landcover classification accuracies of Ikonos and QuickBird images than the other methods, especially with images of urban areas (Akar and Güngör, 2013).+In a comparison of Random Forests ​with three other classification algorithms (Gentle AdaBoost [GAB], Support Vector Machine [SVM] and Maximum Likelihood Classification [MLC]), initial findings indicate that random forests gives higher landcover classification accuracies of Ikonos and QuickBird images than the other methods, especially with images of urban areas (Akar and Güngör, 2013).
  
-Mellor et al. (2013) used random forests ​to classify landcover into forested and non-forested classes using Landsat TM and MODIS imagery of the Australian state of Victoria. ​ They found that using random forests for landcover classification yielded an accuracy of 96% (ϰ = 0.91).+Mellor et al. (2013) used Random Forests ​to classify landcover into forested and non-forested classes using Landsat TM and MODIS imagery of the Australian state of Victoria. ​ They found that using random forests for landcover classification yielded an accuracy of 96% (ϰ = 0.91).
  
-Immitzer, Atzberger, and Koukal (2012) used random forests ​to classify tree species using WorldView-2 images of sunlit crowns in a temperate forest in Austria. ​ They concluded that random forests ​was able to classify tree species with an accuracy of approximately 82%; Random Forests was better able to identify some tree species than others.+Immitzer, Atzberger, and Koukal (2012) used Random Forests ​to classify tree species using WorldView-2 images of sunlit crowns in a temperate forest in Austria. ​ They concluded that Random Forests ​was able to classify tree species with an accuracy of approximately 82%; Random Forests was better able to identify some tree species than others.
  
 ===== Limitations ===== ===== Limitations =====
remote_sensing_methods/random_forests.txt · Last modified: 2013/10/16 13:02 by jgh