Weather-Aware Crop Watering Recommendation System Using Deep Learning

Abstract

The rapid growth of modern agriculture demands intelligent systems to optimize water usage and improve crop productivity, as traditional irrigation methods often lead to inefficient resource utilization and inconsistent crop monitoring. This paper aims to develop a Weather-Aware Crop Watering Recommendation System using deep learning to provide accurate, real-time cultivation guidance. The proposed system utilizes image-based crop classification through YOLO and Convolutional Neural Networks (CNN), combined with trained weather data and internal crop growth stage prediction to generate precise recommendations. The system processes user-uploaded images and outputs crop identification with confidence scores, along with detailed suggestions such as soil type, season, temperature range, water requirements, fertilizer usage, and expected yield. Experimental results demonstrate high prediction accuracy (above 95%) and improved decision-making efficiency, enabling optimized irrigation and reduced water wastage. The proposed approach provides a scalable and intelligent solution for precision agriculture, enhancing sustainable farming practices and supporting data-driven agricultural management in real-world environments.

Introduction

The increasing need for efficient agriculture has highlighted the limitations of traditional farming methods, which rely on manual observation and fixed irrigation schedules. These approaches often result in inefficient water use, poor crop monitoring, and reduced productivity, especially under changing environmental conditions.

To address this, the study proposes a Weather-Aware Crop Watering Recommendation System that uses deep learning and image processing for intelligent farming. The system employs models like CNN and YOLO to analyze crop images, identify crop type, and determine growth stage with high accuracy. It then combines this information with weather data to generate personalized recommendations, including irrigation needs, soil type, temperature range, fertilizer use, and expected yield.

Unlike existing systems that rely only on numerical or weather data, this approach integrates image-based crop analysis with environmental awareness, enabling real-time and adaptive decision-making. It also includes a performance evaluation module to compare model accuracy.

The methodology involves data collection (images and weather data), preprocessing, feature extraction, deep learning-based prediction, and a recommendation engine. The modular system design ensures scalability, real-time processing, and efficient resource utilization.

Overall, the proposed system provides a comprehensive, intelligent, and sustainable solution for precision agriculture by optimizing water usage, improving crop health, and supporting data-driven farming decisions.

Conclusion

This paper presented a Weather-Aware Crop Watering Recommendation System using deep learning techniques to enhance agricultural decision-making. The proposed system successfully integrates image-based crop classification using YOLO and Convolutional Neural Networks (CNN) with environmental data to provide accurate and real-time cultivation recommendations. By identifying the crop type, predicting the growth stage, and incorporating weather conditions, the system generates detailed guidance on soil type, season, temperature, water requirements, fertilizer usage, and expected yield. The experimental results demonstrate that the system achieves high prediction accuracy and provides reliable recommendations, thereby improving irrigation efficiency and reducing water wastage. The integration of weather-aware decision-making further strengthens the system’s ability to support sustainable and precision farming practices. In future work, the system can be enhanced by incorporating real-time weather APIs, expanding the dataset to include a wider range of crops, and utilizing advanced deep learning models to further improve accuracy and scalability. Overall, the proposed system offers a practical and intelligent solution for modern agriculture, contributing to efficient resource management and improved crop productivity.

References

[1] Morchid, A. Elbasri, Z. Oughannou, H. Qjidaa, R. El Alami, and B. Bossoufi, “An innovative smart irrigation using embedded and regression-based machine learning technologies for improving water security and sustainability,” IEEE Access, vol. 13, pp. 145230–145242, 2025. [2] N. Lachgar, M. Essabbar, H. Saikouk, and A. El Hilali Alaoui, “Development of an expert system for precision irrigation: Knowledge modeling approach,” IEEE Access, vol. 13, pp. 165623–165644, 2025. [3] N. Yadav, “IoT-based smart irrigation system using weather forecasting,” International Journal of Science and Research, vol. 13, no. 6, pp. 930–936, 2024. [4] J. Yu, Q. Qu, S. Peng, and X. Wei, “Deep learning for intelligent irrigation decision-making: A review,” Agricultural Water Management, vol. 320, 2025. [5] T. Singh, R. Kumar, and P. Sharma, “Deep learning applications in forecasting agricultural water demand under climate variability,” International Journal of Environmental Sciences, vol. 14, pp. 7078–7084, 2024.

Copyright

Copyright © 2026 A. Askarunisa, K. Dharshini, A. Iswarya. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.