Crops Classification Using Machine Learning And Google Earth Engine

The primary objective of this study was to address the critical challenge of obtaining accurate information regarding the spatial distribution and classification of crops in agricultural areas. The aim was to enhance agricultural decision-making and management, especially in regions with limited wat...

Full description

Saved in:

Bibliographic Details
Published in:	2023 14th International Conference on Intelligent Systems: Theories and Applications (SITA) pp. 1 - 8
Main Authors:	Achahboun, Chakir, Chikhaoui, Mohamed, Naimi, Mustapha, Bellafkih, Mostafa
Format:	Conference Proceeding
Language:	English
Published:	IEEE 22-11-2023
Subjects:	Classification algorithms Crops Decision making image classification Internet Machine learning Machine learning algorithms normalized difference vegetation index spatial analysis Support vector machines
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	The primary objective of this study was to address the critical challenge of obtaining accurate information regarding the spatial distribution and classification of crops in agricultural areas. The aim was to enhance agricultural decision-making and management, especially in regions with limited water resources, where crop productivity and sustainability are crucial for achieving food security and sustainable development. To achieve this objective, this study utilized machine-learning classification algorithms in conjunction with Landsat and Sentinel satellite imagery. The classification of different crops was based on Normalized Difference Vegetation Index (NDVI) phenology . The classification process and post-processing were conducted using the Google Earth Engine (GEE) platform , as well as utilizing Python, and scikit-learn library. Ground-truth data provided by local experts, along with the EUROMAP 2018 dataset in South Spain, were used to label the classification results . The findings of this study demonstrated a classification accuracy of 72% for certain crop types, indicating significant implications for sustainable agricultural practices and land use planning.
DOI:	10.1109/SITA60746.2023.10373760