Random Forest Variable Importance Spectral Indices Scheme for Burnt Forest Recovery Monitoring—Multilevel RF-VIMP

Abstract

Burnt forest recovery is normally monitored with a time-series analysis of satellite data because of its proficiency for large observation areas. Traditional methods, such as linear correlation plotting, have been proven to be effective, as forest recovery naturally increases with time. However, these methods are complicated and time consuming when increasing the number of observed parameters. In this work, we present a random forest variable importance (RF-VIMP) scheme called multilevel RF-VIMP to compare and assess the relationship between 36 spectral indices (parameters) of burnt boreal forest recovery in the Great Xing'an Mountain, China. Six Landsat images were acquired in the same month 0, 1, 4, 14, 16, and 20 years after a fire, and 39,380 fixed-location samples were then extracted to calculate the effectiveness of the 36 parameters. Consequently, the proposed method was applied to find correlations between the forest recovery indices. The experiment showed that the proposed method is suitable for explaining the efficacy of those spectral indices in terms of discrimination and trend analysis, and for showing the satellite data and forest succession dynamics when applied in a time series. The results suggest that the tasseled cap transformation wetness, brightness, and the shortwave infrared bands (both 1 and 2) perform better than other indices for both classification and monitoring.

Forest Stand Species Mapping Using the Sentinel-2 Time Series

“…In the SWIR region reflectance, the water absorption features are important [2,64]. For example, SWIR bands are used as key indicators in forest recovery studies [51].…”

“…In MDA, input variable values are randomly permuted, then the changes in predicted accuracy is measured, while MDG measures the reduction of Gini Impurity metric by a variable for a particular class [47][48][49][50]. In the majority of studies, MDA is used [47] because it is considered as more straightforward, reliable, and easier to understand [51]. However, Behnamian et al [52] claimed that MDG is slightly more stable.…”

The Comparison of Different Methods of Texture Analysis for Their Efficacy for Land Use Classification in Satellite Imagery

“…In addition, for the classification of the GLCM results, the random forest variables importance analysis was performed. Variable importance is calculated based on out-of-bag accuracy and signifies the importance of the respective variable; a high value means a high importance of the variable for the entire random forest model and vice versa [35,39]. The GLCM data set is the only set of utilized data (in the section on texture images) with images of a qualitatively different nature; different images present different features, which in turn are referring to other aspects of the image's texture.…”

Scattering Feature Set Optimization and Polarimetric SAR Classification Using Object-Oriented RF-SFS Algorithm in Coastal Wetlands

“…In the process of random forest model training, the importance of each variable can be calculated [32][33][34]. There are two types of importance measures: mean decrease impurity and mean decrease accuracy (MDA).…”

“…The random forest algorithm not only realizes the classification of remote sensing images, but can also play an important role in feature selection. In the process of random forest modeling, the importance of different feature variables can be calculated and the efficiency of variables in the classification can be quantitatively compared [32]. Most of the current related research focuses on optical remote sensing data or ordinary SAR data [33,34], but there are few studies focusing on polarimetric SAR data and polarimetric scattering features, especially in wetlands monitoring.…”