Robust Multi-Sensor Consensus Plant Disease Detection Using the Choquet Integral

Area of Science:

• Agricultural Technology
• Artificial Intelligence in Agriculture
• Plant Pathology

Background:

• Artificial Intelligence (AI) techniques are increasingly used to enhance sustainable development in agriculture.
• AI facilitates decision-making in the agri-food industry, particularly for automatic plant disease detection.
• Deep learning models analyze plant images for early disease detection, preventing propagation.

Purpose of the Study:

• To design an autonomous Edge-AI device for automatic plant disease detection.
• To incorporate hardware and software for analyzing plant leaf images.
• To improve disease detection accuracy and robustness through data fusion.

Main Methods:

• Development of an Edge-AI device with integrated hardware and software.
• Automatic detection of plant diseases using a set of plant leaf images.
• Implementation of data fusion techniques to enhance classification.

Main Results:

• The Edge-AI device successfully detects plant diseases from leaf images.
• Data fusion techniques significantly improved the classification process.
• The device demonstrated increased robustness in classifying plant diseases.

Conclusions:

• The developed Edge-AI device offers an effective solution for autonomous plant disease detection.
• This technology supports early disease identification, crucial for sustainable agricultural practices.
• The device's robustness contributes to more reliable disease management in the agri-food sector.