Applying Random Forest Classification to Map Land Use/Land Cover Using Landsat 8 Oli

Abstract

<p><strong>Abstract.</strong> This study used the Random Forest classifier (RF) running in R environment to map Land use/Land cover (LULC) of Dak Lak province in Vietnam based on the Landsat 8 OLI. The values of two RF parameters of ntree (number of tree) and mtry (the number of variables used to split at each node) were tested and compared. In current study the best results indicate the number of suitable decision trees involved in the classification process is 300 (ntree), and the suitable number of variables used to split at each node is 4 variables (mtry). These parameters were used to classify 7 bands multi-spectral resolution from 1–7 of Landsat 8 into ten classes of LULC including natural broad-leaved evergreen, semi-evergreen, dipterocarp deciduous forest, plantation forest, rubber, coffee land, crop land, barren land, residential area and water surface. The overall accuracy of 90.32 % with Kappa coefficient of 0.8434 was found in this case.</p>

Land Use and Land Cover Scenarios for Optimum Water Yield and Sediment Retention Ecosystem Services in Klong U-Tapao Watershed, Songkhla, Thailand

“…Additionally. These accuracy results are consistent with previous studies of [67][68][69][70]. However, to apply the RF model under the EnMap BOX software, users are required to observe the preliminary LULC result and add more training sample points to increase the accuracy of classification as mentioned by Reference [71].…”

Land Use/Land Cover Mapping Using Multitemporal Sentinel-2 Imagery and Four Classification Methods—A Case Study from Dak Nong, Vietnam

“…The optimal parameters were determined based on classification error. For the IMG 2-SVM combination and the IMG 4-SVM combination, the optimal C of 2 3 and Gamma of 0.1 produced classification errors of 0.2283 and 0.1525, respectively, while for the IMG 1-SVM and IMG 3-SVM combinations, the optimal C of 2 5 and 2 6 , both with Gamma of 0.1, produced classification errors of 0.2212 and 0.2145, respectively.…”

“…In the last decade, non-parametric methods including support vector machine (SVM) [35][36][37], k-nearest neighbors (k-NN) [38,39], and random forests (RF) [40][41][42] have gained attention for remote sensing-based land cover classification. However, both SVM and RF require the selection of values for multiple parameters that affect their efficacy, and both are computationally intensive [6,35]. For k-NN, Naidoo et al (2012) [43] reported difficulty in selecting the optimal value of k and that the genetic algorithms recommended for optimization can be computationally intensive [44,45].…”

“…Bourgoin et al (in press) [53] used RF with both Landsat and Sentinel-2 data for multiple land cover and land cover change classes and reported an overall accuracy of 0.81. Nguyen et al (2018) [6] used RF and Landsat data for 10 classes including multiple forest classes in Vietnam. Overall accuracy was approximately 0.90.…”

“…For this study, MDA values were investigated to determine the importance of variables. Nguyen et al (2018) [6] indicated that within the range 1 ≤ ntree ≤ 500, ntree = 300 produced the best fit. In addition, Breiman (2001) [78] stated that using more than the required number of trees may be unnecessary, albeit not harmful, because the relationship between accuracy and ntree is asymptotic.…”