Exploring Capabilities of Sentinel-2 for Vegetation Mapping Using Random Forest

Abstract

ABSTRACT:Accurate vegetation mapping is essential for monitoring crop and sustainable agricultural practice. This study aims to explore the capabilities of Sentinel-2 data over Landsat-8 Operational Land Imager (OLI) data for vegetation mapping. Two combination of Sentinel-2 dataset have been considered, first combination is 4-band dataset at 10m resolution which consists of NIR, R, G and B bands, while second combination is generated by stacking 4 bands having 10m resolution along with other six sharpened bands using Gram-Schmidt algorithm. For Landsat-8 OLI dataset, six multispectral bands have been pan-sharpened to have a spatial resolution of 15m using Gram-Schmidt algorithm. Random Forest (RF) and Maximum Likelihood classifier (MLC) have been selected for classification of images. It is found that, overall accuracy achieved by RF for 4-band, 10-band dataset of Sentinel-2 and Landsat-8 OLI are 88.38%, 90.05% and 86.68% respectively. While, MLC give an overall accuracy of 85.12%, 87.14% and 83.56% for 4-band, 10-band Sentinel and Landsat-8 OLI respectively. Results shown that 10-band Sentinel-2 dataset gives highest accuracy and shows a rise of 3.37% for RF and 3.58% for MLC compared to Landsat-8 OLI. However, all the classes show significant improvement in accuracy but a major rise in accuracy is observed for Sugarcane, Wheat and Fodder for Sentinel 10-band imagery. This study substantiates the fact that Sentinel-2 data can be utilized for mapping of vegetation with a good degree of accuracy when compared to Landsat-8 OLI specifically when objective is to map a sub class of vegetation.

Assessment of Sentinel-2 Satellite Images and Random Forest Classifier for Rainforest Mapping in Gabon

“…The authors mapped two forest types (evergreen and moist semi-deciduous forest) in south Ghana applying the SVM method to Sentinel-2 and hyperspectral data and achieved an overall accuracy of 92%. A study by Saini et al [33] demonstrated the superiority of the Sentinel-2 data over the Landsat-8 data in vegetation type classification in India. They tested both datasets for mapping vegetation types including wheat, fodder, trees, fallow land sugarcane, water, other crops built up and sandy area using the RF approach.…”

Integrating Vegetation Indices and Spectral Features for Vegetation Mapping from Multispectral Satellite Imagery Using AdaBoost and Random Forest Machine Learning Classifiers

“…In recent years, with the availability and accessibility of remote sensing data, a huge variety of applications like crop type classification, vegetation mapping, forestry, precision agriculture, landslide susceptibility mapping, built up extraction, etc. have attracted the attention of multidisciplinary researchers [1][2][3][4][5][6][7][8]. Vegetation mapping is one of the essential application needs which has to be addressed effectively for overall environmental monitoring [8].…”

“…Vegetation mapping is one of the essential application needs which has to be addressed effectively for overall environmental monitoring [8]. The utilization of remotely sensed data is the optimal way for vegetation mapping because of the free availability of medium and coarser spatial resolution data, having different spatial and spectral properties, cost effective and less time consuming in comparison to traditional field survey methods [7,8]. Quantifying vegetation provides a valuable information for socio-economic applications.…”

“…In the past decade, various Machine Learning techniques such as Artificial Neural Network (ANN), k-Nearest Neighbor (k-NN), Random Forest, Support Vector Machine (SVM), Rotation Forest, various boosting techniques (AdaBoost, Schotostic Gradient Boosting etc.) have attracted considerable interest from multidisciplinary researchers as well as within the remote sensing community because these classifiers provide high classification accuracy and are robust to noise [2][3][4][5][6][7][8]. The ensemble classifier is a kind of Machine Learning method which consists of a number of base classifiers and combines their prediction by means of some voting scheme or mechanism.…”

“…The basic concept behind an ensemble classifier makes the assumption that a combination of multiple base classifiers provides better prediction results as compared to a single classifier [10][11][12]. Random Forest is one popular ensemble method used for a variety of remote sensing applications like landslide susceptibility mapping [5], vegetation mapping [2,4,7,12], ecotope mapping [1], land cover classification [3], etc. AdaBoost is a popular boosting technique proposed by Freund and Schapire [13], and this algorithm enhances the performance of base learner or weak models by converting them into the strong learner or models [8,14].…”