Author(s) :

Fatima, Maria Anjum, Muhammad Hamza Ibrahim, Muhammad Rafiq, Ali Raza.

Volume/Issue :

Abstract :

Global crop output is severely hampered by pests and weeds, which frequently result in significant yield losses and a growing need for chemical control methods. Conventional monitoring techniques require a significant amount of time and effort, often resulting in the detection of infestations too late, which can lead to extensive damage and the overuse of pesticides. New developments in artificial intelligence (AI), machine learning (ML), and the Internet of Things (IoT) offer creative ways to identify, monitor, and forecast insect and weed outbreaks early. While deep learning and convolutional and neural networks allow for accurate image-based identification of pest and weed species, neural networks and decision tree predictive models predict insect population dynamics with high accuracy. in addition to increasing accuracy, these technologies optimize the use of herbicides, fertilizers, and water, which lower expenses and encourages sustainable farming. Despite ongoing obstacles including data unpredictability, acceptance hurdles, and technological constraints, interoperating AI and ML into agriculture has enormous potential for intelligent, environmentally friendly crop protection.

No. of Downloads :

0