A New Approach for Mapping Land Use / Land Cover Using Google Earth Engine: A Comparison of Composition Images

Abstract

Abstract. In view of the increase in human activities, climate change and related hazards, land use and land cover (LULC) mapping is becoming a fundamental part of the process of any development or hazard prevention project. From this perspective, we propose a new approach for mapping LULC using Machine learning algorithms by comparing the result of five composition methods based on Google Earth Engine in the city of Tetouan - Morocco. To achieve this goal, considering the Sentinel S2 L2 imageries as a source data , five datasets were derived to make the classification generating by aggregating functions (median , mean , max , min and mode). Then based on the very high resolution (VHR) satellite images provided by Google Earth comes the next step that involves selecting samples that are divided into five classes (barren land, water surface, vegetation, forest, and urban areas), which will be further split into two parts: 70% as a training data -used to feed the machine learning algorithms (support vector machine (SVM), random forest (RF) and classification and regression trees (CART))- and 30% as a testing data for evaluating the models using accuracy assessments. The results for all datasets indicate that the SVM algorithm has the highest accuracy and its performance is better than the other algorithms (RF and CART). The average overall accuracy of SVM , RF , and CART was 87.99% , 87.81% and 84.72% , respectively. Furthermore, for each algorithm, the comparison between the results of the different composites indicates that the use of the mean composite is the most suitable for LULC mapping. Finally, GEE has proven to be an effective and rapid method for LULC mapping, especially with the use of compositional imagery that can assist decision makers in future planning or risk prevention.

Comparison of machine and deep learning algorithms using Google Earth Engine and Python for land classifications

“…These datasets were employed to map the region of Sukkur, Pakistan, with the aim of delineating the area of each zone or category (Water bodies, Built-up, Barren Land, and Vegetation) and assessing the performance of each classifier utilized in this research. In accordance with our examination of previous studies (Yang et al, 2021;Nelson et al, 2022;Prasad et al, 2022;Sertel et al, 2022;Yang et al, 2022;Sellami and Rhinane, 2023;Zhao et al, 2023;Yari Hesar et al, 2024). From the recent studies, we discovered that the methodology employed in our experimentation could be adapted for mapping and evaluating various areas in different countries or cities.…”

Supervised Machine Learning Algorithms for Land Cover Classification in Casablanca, Morocco

“…Integrating the results of previous studies [3,19,22,26,[37][38][39][40][41][42] and our own research, the main achievement of this study lies in the successful adaptation of the methodology to land use mapping and evaluation in different geographical contexts.…”

“…Previous research [3,[21][22][23], particularly in European and Asian regions, has extensively utilized GEE and various algorithms for land cover classification, providing valuable insights into the effectiveness of remote sensing technologies and machine learning algorithms in identifying land cover dynamics. The research community has employed GEE to examine diverse landscapes and land use patterns, contributing to an understanding of environmental changes and ecosystem dynamics.…”

Comparison of Machine Learning Methods for Satellite Image Classification: A Case Study of Casablanca Using Landsat Imagery and Google Earth Engine

“…By incorporating these indices, the study aimed to improve the classification accuracy and achieve more precise identification of land cover categories. Based on previous studies [9,14,16,[19][20][21][22][23][46][47][48][49][50][51][52] and our own research, we found that the methodology em-ployed in our experiment can be adapted to map and evaluate different regions in various countries or cities, as this methodology is not oriented solely toward Casablanca city. The main contribution of this research has been the successful adaptation of the method to map and evaluate land use in other regions or countries.…”

“…Although the literature review [14,[22][23][24] indicates extensive research on land cover classification using GEE and various algorithms in European and Asian regions, there is little specific research for Morocco. This review is an opportunity for researchers to ex-plore land cover classification in Casablanca, Morocco, using supervised and unsupervised algorithms by exploiting the advantages of the GEE platform [13,25] .…”