Published January 1, 2026 | Version v1

Geographical Remote Data Analysis for Agriculture Crop Production Prediction

Description

Accurate estimation of crop water requirements is essential for efficient irrigation management and sustainable utilization of available water resources. This paper proposes a crop water requirement prediction model based on a Teacher Learning-Based Optimization Genetic Algorithm (TLBO-GA). The proposed approach utilizes Normalized Difference Vegetation Index (NDVI), Vegetation Condition Index (VCI), and Standardized Precipitation Index (SPI) as significant features for representing vegetation and climatic conditions. During optimization, the teacher and learner phases improve candidate cluster centers through knowledge sharing, while genetic crossover and mutation enhance population diversity and reduce premature convergence. The optimized representative features are subsequently provided to the EBPNN for training and prediction. Experimental analysis using real-world data from Madhya Pradesh, India, demonstrates that the proposed model achieves an average prediction accuracy of 97.83%. The results demonstrate that TLBO-GA-based feature optimization provides an effective and computationally efficient solution for crop water requirement prediction.

Files

IJSET_V14_issue4_227.pdf

Files (635.3 kB)

Name Size
md5:a5855aeb2a4d30a39831b080b24615a2
635.3 kB Preview Download

Additional details