Evaluation of Machine Learning Models for Mapping Food Crops using Sentinel-2A Imagery in West Java, Indonesia

Aug 3, 2026

Upload by Levi

Food crop distribution patterns and locations are essential for monitoring agricultural sustainability and strengthening food security. Machine-learning-based plant classification is widely used to detect cropland areas, yet challenges remain in mapping plant types and estimating crop area efficiently. This study evaluates SVM, Random Forest, and CART for mapping and calculating rice crop area in West Java, Indonesia, using Sentinel-2A time-series features (NDVI and EVI) in Google Earth Engine.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages17

Reading Time43 min

AccessFree to download

Introduction

Food security needs and agricultural monitoring

Remote sensing for crop mapping

Challenges: data quality vs cost and resolution limits

Sentinel-2A as an alternative for mapping

Agronomic factors affecting classification

Evaluation of Machine Learning Models for Mapping Food Crops using Sentinel-2A Imagery in West Java, Indonesia Riki Ridwana1,2, Muhammad Kamal3, Sanjiwana Arjasakusuma3, and Muh Fiqri Abdi Rabbi4 1Mapping Survey and Geographic Information Study Program, Faculty of Social Sciences Education, Universitas Pendidikan Indonesia, Jl. Dr. Setiabudhi No. 229, Bandung, Indonesia 2Doctoral Program in Geography, Faculty of Geography, Universitas Gadjah Mada, Yogyakarta, Indonesia 3Departement of Geography Information Science, Faculty of Geography, Universitas Gadjah Mada, Bulaksumur, Yogyakarta 55281, Indonesia 4Geography Information Science Study Program, Faculty of Social Sciences Education, Universitas Pendidikan Indonesia, Jawa Barat, Indonesia. Abstract. Data on the distribution patterns and locations of food crops are crucial for monitoring and controlling the sustainability of agricultural resources and guaranteeing food security. Plant classification based on machine learning has been widely used to detect food crop areas. However, there are still challenges in mapping plant types and plant area effectively and efficiently. The aim of this research is to evaluate machine learning models in mapping and calculating the area of food crops (rice) in West Java Province, Indonesia. Google Earth Engine is used in this study as a big data cloud computing platform for remote sensing. Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI) Sentinel- 2A imagery is utilized to employ time series data as input characteristics for the three most popular machine learning models: Support Vector Machine (SVM), Random Forest (RF), and Classification and Regression Trees (CART) . The research results show that the three machine learning models are able to map and calculate the area of food crops in West Java, Indonesia. The RF algorithm produces the highest overall accuracy rate (98.51%) and is the fastest in the accuracy assessment and image classification process compared to the SVM and CART algorithms. 1 Introduction The quick expansion of world urbanization and population is pressing for food production from limited agricultural resources to maintain food security [1–4] . In order to satisfy the demands for food supply, a large number of sophisticated agricultural systems have been developed with careful monitoring and management of agricultural resources to boost efficiency and crop yields [5-7]. Monitoring and managing sustainable agricultural resources greatly benefits from knowing the locations and patterns of food crop distribution [8-11] . This information is used as a basis for policy makers in estimating harvest yields [12, 13], © The Authors, published by EDP Sciences. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0 ([https://creativecommons.org/licenses/by/4.0/](https://creativecommons.org/licenses/by/4.0/)). Estimated crop specific water consumption [14-16], disaster assessment [17, 18], and ensuring balanced regional food security [19, 20]. Consequently, crop maps are a trustworthy tool for sustainable agriculture management that decision makers may rely on. So far, spatial information on plant types and planting areas still uses traditional methods, namely based on field surveys, area unit frameworks, and statistical reports, which take a lot of time, are labor intensive, cost a lot of money, and lack timely data updates [21–23] . On the other hand, Nowadays, plant kinds may be reliably mapped using remote sensing technology, which offers broad coverage capabilities, quick data collecting, and cost and human resource savings [24-26] . Although methods for mapping plant types in the last two decades have been developed using different remote sensing data [27–29], There are still gaps and challenges to develop remote sensing technology. One of these challenges is the limitation between ima

FAQs

What is the goal of this research on food crops mapping?

To evaluate machine learning models for mapping and calculating the area of food crops, specifically rice, in West Java, Indonesia.

Which machine learning models are compared in the study?

Support Vector Machine (SVM), Random Forest (RF), and Classification and Regression Trees (CART) are used and compared for mapping crop areas.

Why are NDVI and EVI from Sentinel-2A imagery used?

NDVI and EVI Sentinel-2A time series are used as input characteristics to capture vegetation signals that support plant type mapping and area estimation.

Aug 3, 2026

Upload by Levi