Comparing six pixel-wise classifiers for tropical rural land cover mapping using four forms of fully polarimetric SAR data
Citations
83 citations
Cites methods from "Comparing six pixel-wise classifier..."
...The GEOBIA paradigm coupled with the use of machine learning classification algorithms is currently considered an excellent “first-choice approach” for the classification of forest tree species and the general derivation of forest information [28,31,41,55,63]....
[...]
69 citations
Cites background from "Comparing six pixel-wise classifier..."
...On the other hand, RF has generally been found to be robust to parameter settings (RodriguezGaliano et al. 2012; Trisasongko et al. 2017; Trisasongko and Paull 2019; Maxwell, Warner, and Fang 2018)....
[...]
...Conversely, Foody and Mathur (2004) and Trisasongko et al. (2017) found larger impact on classification accuracies, with differences up to 20% when varying the SVM parameters....
[...]
42 citations
41 citations
30 citations
References
272,030 citations
[...]
79,257 citations
"Comparing six pixel-wise classifier..." refers background in this paper
...This approach, known as Random Forests (RFs), was introduced by Breiman (2001) and has been exploited in remote sensing....
[...]
...This approach, known as Random Forests (RFs), was introduced by Breiman (2001) and has been exploited in remote sensing. For instance, Clewley et al. (2015) reported that significant accuracy was...
[...]
40,147 citations
"Comparing six pixel-wise classifier..." refers methods in this paper
...Support Vector Machine (SVM) Another classifier that has been widely utilized in remote sensing is SVM (Vapnik 2000)....
[...]
17,764 citations
4,319 citations
"Comparing six pixel-wise classifier..." refers background in this paper
...However, for the sake of efficiency, RPROP may be the primary choice as the backpropagation algorithm is usually computationally inefficient (Riedmiller and Braun 1993)....
[...]