Forest Stand Types Classification Using Tree-Based Algorithms and SPOT-HRG Data

Forest types mapping is one of the most necessary elements in forest management and silviculture treatments. traditional methods such as field surveys are time-consuming and cost-intensive. improving satellite data sources and classification methods offer new opportunities for obtaining more accurate forest biophysical maps. this research compares performance of three non-parametric and tree-based algorithms i.e. the classification and regression tree (cart), boosting regression tree (brt) and random forest (rf) for general forest type mapping using semi high resolution of spot-hrg data. using systematic random sampling design in a small area of the hyrcanian forests, tree and shrub species were registered in 150 sample plots. naming of the general forest types in sample plots were done based on frequency of dominant species. after geometric and atmospheric corrections of sopt-hrg data, suitable image processing transformations were applied to main bands to produce general vegetation indices and principal components. three nonparametric algorithms performed the wall-to-wall forest type classification. the forest type maps were assessed using unused test plots. results shows that rf compared to the other two algorithms with overall accuracy of 70% and kappa coefficient of 0.63 could better classify the forest stand types, while the cart method had the lowest accuracy with overall accuracy of 60% and kappa coefficient of 0.51. performance results of the brt classifier were slightly similar to rf classifier.