📅 July 2023
🧑 S. T. Sasidharan, D. Latini, M. Ivanovici, G. Schiavon, K. Thangavel, F. Del Frate
This poster presents a method for improving weed control in agriculture through the use of Unmanned Aerial Vehicles (UAVs) equipped with cameras and Deep Neural Networks (DNN). The objective is to detect weeds in real-time using Edge Artificial Intelligence (AI), where the processing is done directly on the UAVs without needing external servers. This enables real-time actions, such as the immediate eradication of weeds by robots operating on the ground.
Key points include:
- Weeds compete with crops for nutrients, water, and light, leading to reduced yields and quality, while also harboring pests and diseases.
- Edge AI allows real-time data processing on UAVs, reducing reliance on manual labor and the use of harmful chemicals.
- The preliminary results show successful weed detection using onboard hardware accelerators, and the methodology includes a workflow for this process.
This technology aims to increase agricultural productivity while minimizing chemical use and manual labor.
https://ai4agri.unitbv.ro/wp-content/uploads/2024/04/S_T_Sasidharan_EARSeL_2023.pdf