Smart IoT-driven precision agriculture: Land mapping, crop prediction, and irrigation system
Top Cited Papers
Open Access
- 18 March 2025
- journal article
- research article
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 20 (3) , e0319268
Abstract
As the world population is increasing day by day, so is the need for more advanced automated precision agriculture to meet the increasing demands for food while decreasing labor work and saving water for crops. Recently, there have been many studies done in this field, but very few discuss implementing smart technologies to present a combined sustainable farming system. In this article, we present a complete integrated design of a smart IoT-based suitable agricultural land and crop selection, along with an irrigation system using agricultural mapping, machine learning, and fuzzy logic for precision agriculture. Multi-spectral band images from Landsat-8 satellite images of a chosen land are employed from USGS Earth Resources Observation and Science (EROS) Center for extracting indices that are used for agricultural analysis, determining the vegetation index, water index, and salinity index of that land using K-means. Furthermore, crop yield is predicted using Linear Regression and Random Forest, achieving accuracies of 93.49% and 95.87%, respectively, while using RMSE (Root Mean Squared Error) as the loss function. The LSTM model is used for healthy vegetation area forecasting highlighting the changes of the vegetation area over time. Such analysis helps to decide whether that land is suitable for farming or not. Multiple soil-parameter measuring sensors are used to identify suitable crop and fertilizer requirements for that land using IoT and machine learning. The ML model-based crop prediction showed 97.35% accuracy utilizing random forest algorithm. Finally, a fuzzy logic-based solar-powered irrigation system is used to monitor the water requirements of those crops and irrigate them according to their needs. The experimental results demonstrated that fuzzy logic has faster calibration rate of 66.23% and helps to save around 61% water in comparison to average logic algorithm. The implementation of a fuzzy logic algorithm significantly optimized water usage compared to traditional manual irrigation methods. These findings highlight the effectiveness of advanced computational techniques in enhancing agricultural practices and resource management.Keywords
This publication has 41 references indexed in Scilit:
- CropCast: Harvesting the future with interfused machine learning and advanced stacking ensemble for precise crop predictionKuwait Journal of Science, 2023
- Zoning irrigation smart system based on fuzzy control technology and IoT for water and energy savingJournal of Cleaner Production, 2021
- Satellite-based soybean yield forecast: Integrating machine learning and weather data for improving crop yield prediction in southern BrazilAgricultural and Forest Meteorology, 2020
- Fuzzy Logic based Smart Irrigation System using Internet of ThingsJournal of Cleaner Production, 2020
- Remote sensing for agricultural applications: A meta-reviewRemote Sensing of Environment, 2020
- Crop yield prediction with deep convolutional neural networksComputers and Electronics in Agriculture, 2019
- Temporal relationships between time series CHIRPS-rainfall estimation and eMODIS-NDVI satellite images in Amhara Region, Ethiopia2019
- Analysis of reference evapotranspiration (ET0) trends under climate change in Bangladesh using observed and CMIP5 data setsMeteorology and Atmospheric Physics, 2018
- Mapping cultivable land from satellite imagery with clustering algorithmsInternational Journal of Applied Earth Observation and Geoinformation, 2016
- Crop yield prediction under soil salinity using satellite derived vegetation indicesField Crops Research, 2016