Original scientific paper

https://doi.org/10.17559/TV-20241113002127

Enhancing Precision Agriculture with a Novel AI Framework for Early Crop Health Detection

Chandraleka J. ; Department of Computing Technologies, Faculty of Engineering and Technology, School of Computing, SRM Institute of Science and Technology, Kattankulathur, Chennai, 603203, Tamil Nadu, India
Selvaraj P. ; Department of Computing Technologies, Faculty of Engineering and Technology, School of Computing, SRM Institute of Science and Technology, Kattankulathur, Chennai, 603203, Tamil Nadu, India *

* Corresponding author.

Abstract

The Precision agriculture is a groundbreaking approach that leverages advanced technologies to optimize agricultural productivity and sustainability. By integrating data-driven techniques, farmers can monitor and manage their crops more effectively, ensuring better yields and resource efficiency. One of the key challenges in precision agriculture is the early identification of factors that may negatively impact crop health. Microorganisms and environmental stressors can silently affect plants, often remaining undetected until they cause significant damage. This paper presents a novel methodology employing a Logistic Activation function with a Modified Fuzzy-based Convolutional Neural Network (LA-MFCNN) algorithm, designed to enhance the early detection of potential threats to crop health, specifically targeting sugarcane cultivation. Our approach utilizes fuzzy logic principles combined with deep learning techniques to analyze complex data patterns, identifying subtle indicators that may signal emerging issues in crop health. The LA-MFCNN algorithm is specifically engineered to recognize and interpret early warning signs, enabling timely interventions and mitigating potential risks. By leveraging artificial intelligence, this method facilitates more accurate and efficient monitoring, thereby supporting decision-making processes in precision agriculture. The performance of the proposed LA-MFCNN algorithm is rigorously compared against traditional Machine Learning (ML) and Deep Learning (DL) algorithms. Key performance metrics, including accuracy, precision, recall, and F1 score, demonstrate that our approach significantly outperforms existing methods. The results underscore the algorithm's potential to revolutionize precision agriculture by improving crop management strategies and enhancing agricultural productivity. Furthermore, the adaptability of the proposed method allows for its application to various crops, making it a versatile tool for modern agriculture. This research highlights the critical role of advanced AI techniques in transforming traditional farming practices, paving the way for more sustainable and efficient agricultural systems.

Keywords

artificial intelligence; convolutional neural networks; deep learning; fuzzy logic; precision agriculture

Hrčak ID:

335059

URI

https://hrcak.srce.hr/335059

Publication date:

30.8.2025.

Visits: 397 *