Feasibility Study of Water Stress Detection in Plants using a High-Throughput Low-Cost System

The importance of efficient water management for crops and its impact on canopy development is undeniable, where high-throughput systems connected to the cloud have a great potential to aggregate measurements for a cyber-physical analysis. In this work, we develop a vision based high-throughput system to aid the identification and early detection of stress caused by water deficit in maize plants. We show that image processing techniques combined with a near infrared camera allow the distinction between a well watered maize plant and one that has its water supply neglected. Our preliminary results demonstrate the potential viability of applying predictive models for early water stress detection.