Water Quality Classification for Precision Irrigations System Using Machine Learning and Remote Sensing

Abstract

Muhammad Rashid, Shehroz Ejaz, Omer Farooq, Uzma Rafeeq and Abbira Taswar

Estimating water quality has been one of the significant challenges faced by the world in recent decades. Ensuring efficient and sustainable irrigation relies heavily on accurate water quality assessment. Contaminated water can harm soil health, crop yield, and the agricultural ecosystem. Developing precise water quality classification models is crucial, especially with the increasing demand for precision irrigation systems. This study proposes a water quality prediction model using Principal Component Regression (PCR) and Gradient Boosting Classifier (GBC). The Water Quality Index (WQI) is calculated, and Principal Component Analysis (PCA) extracts dominant water quality parameters. Several regression algorithms, including Support Vector Regression (SVR), are applied to predict WQI values. Experimental results show that the PCR model with SVR achieves 95% prediction accuracy. The Gradient Boosting Classifier achieves 100% accuracy in classifying water quality levels. The proposed approach enhances prediction reliability while reducing required parameters.

PDF