Skip to main content

Table 8 Evaluation of the accuracy of RF classification for the different classes and set of variables

From: Assessing a novel modelling approach with high resolution UAV imagery for monitoring health status in priority riparian forests

No. of classes considered Overall accuracy for all variables Overall accuracy without
texture variables from NDVI and DSM
Overall accuracy without
variables from DSM
4 classes 67% 60% 59%
3 classes 72% 62% 61%
2 classes 90% 86% 85%