Abstract
Random Forest (RF) is a widely used algorithm for classification of remotely sensed data. Through a case study in peatland classification using LiDAR derivatives, we present an analysis of the effects of input data characteristics on RF classifications (including RF out-of-bag error, independent classification accuracy and class proportion error). Training data selection and specific input variables (i.e., image channels) have a large impact on the overall accuracy of the image classification. High-dimension datasets should be reduced so that only uncorrelated important variables are used in classifications. Despite the fact that RF is an ensemble approach, independent error assessments should be used to evaluate RF results, and iterative classifications are recommended to assess the stability of predicted classes. Results are also shown to be highly sensitive to the size of the training data set. In addition to being as large as possible, the training data sets used in RF classification should also be (a) randomly distributed or created in a manner that allows for the class proportions of the training data to be representative of actual class proportions in the landscape; and (b) should have minimal spatial autocorrelation to improve classification results and to mitigate inflated estimates of RF out-of-bag classification accuracy.
Highlights
Random Forest (RF) is a widely used algorithm for classification of remotely sensed data
The results of this study demonstrate that RF image classification is highly sensitive to training data characteristics, including sample size, class proportions and spatial autocorrelation
We have demonstrated that the results of RF classification can be inconsistent depending on the input variables and strategy for selecting the training data used in classification
Summary
Random Forest (RF) is a widely used algorithm for classification of remotely sensed data. The input variables (i.e., image channels) are randomly selected for building trees These characteristics of the algorithm allow RF to produce an accuracy assessment called “out-of-bag” error (rfOOB error) using the withheld training data as well as measures of variable importance based on the mean decrease in accuracy when a variable is not used in a building a tree. When performing image classification and accuracy assessments, training and validation data should be statistically independent (e.g., not clustered) [14] and representative of the entire landscape [10,12], and there should be abundant training data in all classes [15]. Many different training and validation sampling schemes are used throughout the literature, but without careful scrutiny of each dataset used and the specific assessment method, it may be difficult to compare results of classifications [11,16]. Care must be taken to ensure validation points are drawn from a sample independent of training data to avoid optimistic bias [14]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
