Deep Learning for Stress Plant Phenotyping

This paper presents a deep learning-based plant phenotyping system aimed at reducing the effects of hybrid and heat stress in barley (Rihane variety). The system uses a database of plant images taken by Canon Camera 2000D combined with environmental data, such as temperature and humidity, collected through DHT11 sensors. By leveraging a deep learning model (CNN) implemented on the Collaboratory platform, the system analyzes these images to assess plant stress conditions. Advances in AI and remote sensing allow for precise phenotypic data collection, enabling improved field management. The proposed approach achieved around 90% accuracy, offering a more efficient and cost-effective method for field surveys and plant phenotyping.