In the heart of Punjab, Pakistan, a team of researchers led by Shabeena Naveed from the University of Management and Technology in Sialkot has developed a groundbreaking Android application called AgriSage. This innovative tool is set to revolutionize the agricultural sector by leveraging artificial intelligence and e-commerce to empower farmers and enhance decision-making processes.
AgriSage integrates several advanced technologies, including AI-driven disease detection, weather forecasts, and updates on governmental schemes. The application is designed to support a wide range of users, from farmers and sellers to customers and policymakers. “Our goal was to create a comprehensive platform that addresses the critical challenges faced by the agricultural community, such as timely disease detection and fragmented market access,” Naveed explained.
One of the standout features of AgriSage is its use of two optimized deep learning models for on-device deployment. The first model, based on MobileNetV2, performs binary classification to detect the presence of plants in images. It achieved a perfect classification performance with a precision, recall, and F1-score of 1.00 for both classes. The second model, a convolutional neural network (CNN) designed for disease classification, was trained on the PlantVillage dataset across 38 classes. It achieved a macro average F1-score of 0.8207 and a weighted average F1-score of 0.8703. The optimized TensorFlow Lite version of this model demonstrated an average inference time of 35.6 milliseconds per image, making it suitable for real-time, on-device deployment.
The implications of this research are vast. AgriSage not only provides a robust and scalable platform for AI-powered crop monitoring and disease detection but also delivers real-time agricultural support services. This can lead to improved decision-making and promote sustainable farming practices. “By integrating these technologies, we aim to bridge the gap between farmers and the latest advancements in agricultural science,” Naveed added.
The commercial impacts of AgriSage are equally significant. For the energy sector, the application’s ability to optimize crop monitoring and disease detection can lead to more efficient use of resources, reducing the environmental footprint of agricultural practices. This, in turn, can contribute to the development of a more sustainable and energy-efficient agricultural industry.
Published in the journal ‘Engineering Reports’ (translated to English as ‘Reports on Engineering’), this research highlights the potential of AI and deep learning in transforming the agricultural sector. As the world grapples with the challenges of climate change and food security, innovations like AgriSage offer a glimmer of hope and a path forward. The research team’s work not only shapes future developments in the field but also sets a new standard for integrating technology into agriculture.