FROM SATELLITE TO SOIL: A HIGH-RESOLUTION HYBRID SENSING FRAMEWORK FOR ACCURATE FARM-LEVEL YEILD ESTIMATION AND TRANSPARENT CROP INSURANCE

Authors: Dhiraj Sachin Dahale, Parag Hemraj Patil, Gayatri Sandip Nawale, Chaitanya Pravin Thorat, Isha Singh Rajput, Gauri Balasaheb Chature

DOI: https://doi.org/10.37082/IJIRMPS.v13.i6.232843

Short DOI: https://doi.org/hbdtd9

Country: India

Full-text Research PDF File:   View   |   Download

Abstract: Accurate farm-level crop yield estimation is essential for improved agricultural planning, insurance claims, and quick decision-making. While systems like YES-TECH provide reliable village or Insurance Unit–level estimates, they overlook within-field variability and localized crop damage. This research introduces a high-resolution framework that integrates very-high-resolution satellite imagery, UAV data, and process-based crop models to estimate yields at the farm scale. The system extracts biophysical indicators, generates vegetation indices, and uses a hybrid approach combining crop growth simulations with machine learning to deliver precise predictions. Results show reduced estimation errors, better detection of stress zones, and more accurate post-calamity loss assessments compared to existing unit-level models. The findings demonstrate that high-resolution sensing, paired with robust modelling, can significantly enhance transparency, fairness, and efficiency in crop insurance. The study underscores the potential of digital farm-level yield estimation to support precision agriculture and strengthen data-driven decisions for farmers, insurers, and government bodies.

Keywords: Farm-level yield estimation, remote sensing, crop modelling, machine learning, UAV imagery, precision agriculture.

Paper Id: 232843

Published On: 2025-12-07

Published In: Volume 13, Issue 6, November-December 2025

Share this