Title: Precision agriculture with machine learning: multi-crop identification from remote sensing data

Authors: Khushbu Maurya

Addresses: Computer Engineering Department, Indus University, Ahmedabad, Gujarat, India

Abstract: Accurate identification and classification of multiple crops are essential for efficient agricultural management, productivity assessment, and sustainable land use planning. Leveraging advanced machine learning (ML) techniques combined with remote sensing data provides a powerful approach for precise, large-scale crop monitoring. This study presents a multi-crop identification framework utilising ensemble ML classifiers and multi-source remote sensing imagery, including multispectral, hyperspectral, and radar data. Key steps include spectral and textural feature extraction, vegetation index calculation, and data fusion for improved classification accuracy. Our approach integrates satellite data with machine learning algorithms to distinguish crop types in complex agricultural landscapes, achieving high-resolution mapping and reliable temporal analysis. Results demonstrate the model's capability to discriminate among various crops effectively, highlighting its potential for real-time crop monitoring and land use analysis, ultimately supporting data-driven agricultural decision-making and policy development.

Keywords: multi-crop identification; machine learning; remote sensing; ensemble classifiers; vegetation indices; temporal analysis; precision agriculture.

DOI: 10.1504/IJAITG.2026.153492

International Journal of Agriculture Innovation, Technology and Globalisation, 2026 Vol.5 No.2, pp.153 - 161

Received: 20 Mar 2025
Accepted: 17 Nov 2025
Published online: 11 May 2026 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article