Testing Accuracy of Land Cover Classification Algorithms in the Qilian Mountains Based on GEE Cloud Platform

Abstract

The Qilian Mountains (QLM) are an important ecological barrier in western China. High-precision land cover data products are the basic data for accurately detecting and evaluating the ecological service functions of the QLM. In order to study the land cover in the QLM and performance of different remote sensing classification algorithms for land cover mapping based on the Google Earth Engine (GEE) cloud platform, the higher spatial resolution remote sensing images of Sentinel-1 and Sentinel-2; digital elevation data; and three remote sensing classification algorithms, including the support vector machine (SVM), the classification regression tree (CART), and the random forest (RF) algorithms, were used to perform supervised classification of Sentinel-2 images of the QLM. Furthermore, the results obtained from the classification process were compared and analyzed by using different remote sensing classification algorithms and feature-variable combinations. The results indicated that: (1) the accuracy of the classification results acquired by using different remote sensing classification algorithms were different, and the RF had the highest classification accuracy, followed by the CART and the SVM; (2) the different feature variable combinations had different effects on the overall accuracy (OA) of the classification results and the performance of the identification and classification of the different land cover types; and (3) compared with the existing land cover products for the QLM, the land cover maps obtained in this study had a higher spatial resolution and overall accuracy.

Detecting and distinguishing between apicultural plants using UAV multispectral imaging

“…In this study, we tested how different drone band combinations, referring both to single spectral bands and spectral indices can affect the classification results and which can be considered as the best combination resulting in higher thematic accuracy. According to the literature, existing studies have shown that the use of spectral indices can increase classification accuracy ( Phan, Kuch & Lehnert, 2020 ; Praticò et al, 2021 ; Yang et al, 2021 ). The spectral response of apicultural plants was estimated using six commonly used vegetation indices (Vis).…”

Tracing the Land Use Changes in Ishurdi and Bheramara Upazila, Bangladesh: Application of Google Earth Engine

“…Every decision tree would provide a cataloging choice utilizing the tool of numerous decision tree polling to fix the problem of easy decision tree overfitting and the widely held voting appliance policy to get the ultimate output (Paul et al, 2018;Wu et al, 2018). Some researchers have done significant work on the LU/LC grouping by using the algorithm of RF on the GEE stage, and the outcomes have been excellent (Zhang & Zhang, 2020;Yang et al, 2021;Phan et al, 2020). LULC types were done by directly calling the "ee.smileRandomForest function" from GEE of API, which merely necessities to know two parameters as the classification trees number and the feature variables number that entered at the node splitting time (Liu et al, 2020;Loukika et al, 2021).…”

Mapping Spatiotemporal Changes of Evergreen Forest Patches that are Heritage Sites in Southern Mozambique using Google Earth Engine

“…The most frequently used method to assess the classification performance of satellite images is a confusion matrix containing producer accuracy, consumer accuracy, overall accuracy and kappa coefficient (Yang et al, 2021;Basheer et al, 2022). The remaining 30% of the points (validation points) were applied to compute a confusion matrix comparing all three land-cover maps to their reference points (Basheer et al, 2022;Ochungo et al, 2022).…”