We are looking for motivated Master’s students to join us for the following master thesis projects:

1. AI-Driven Smart Monitoring and Disease Management System for Cowpea Cultivation
2. more will be posted

Contact: Please contact Prof. Shirley Siu by email or pay a visit to the Academic Building E710-4 @ MPU.

2026/2027

AI-Driven Smart Monitoring and Disease Management System for Cowpea Cultivation

Background and Motivation:

The cowpea plant (Vigna unguiculata) is an important legume crop with high nutritional and economic value. However, its productivity is frequently affected by plant diseases attacks.  Certain diseases are caused by bacteria, viruses, and fungi. Traditional field monitoring relies heavily on manual inspection, which is labor-intensive, subjective, and inefficient at scale. Advances in computer vision, edge computing, and precision agriculture provide an opportunity to develop automated, scalable systems for real-time crop health monitoring and disease diagnosis.

Objectives:

The primary goal of this project is to develop a deep learning model and an intelligent system for monitoring cowpea fields and diagnosing leaf disease using image data. Specific objectives include:

• Develop a robust image-based disease detection model for cowpea leaves.
• Integrate field-level monitoring using IoT-enabled imaging devices (e.g., drones or ground sensors).
• Design a decision-support module for disease management recommendations.
• Evaluate system performance under real-world agricultural conditions.

Materials and Methodology:

Data Collection and Curation:

We will acquire a dataset of cowpea leaf images under varying field conditions (healthy and common diseases). These images will be annotated with disease labels (e.g., powdery mildew, downy mildew, root rot), and detailed text description (e.g., conditions of leaf vein, leaf surface, leaf abaxial surface).

Model Development:

• Implement deep learning-based image classification and/or segmentation models (e.g., CNNs, Vision Transformers). 
• Compare architectures for accuracy, efficiency, and generalization. 
• Incorporate transfer learning from pretrained plant disease datasets.

Field Monitoring System:

• Design a prototype system using drones or fixed cameras for periodic image acquisition. 
• Integrate edge or cloud-based inference pipelines for near real-time analysis.

Disease Management Module:

• Develop a rule-based or ML-driven recommendation system linking detected diseases to actionable interventions (e.g., pesticide application, irrigation adjustments). 
• (Optionally) integrate environmental data (temperature, humidity) to improve predictive accuracy.

Evaluation and Validation:

• Assess model performance using metrics such as accuracy, F1-score, and IoU (for segmentation). 
• Validate system usability and reliability in a pilot pea field deployment.

Expected Outcomes:

• A validated AI model capable of accurately diagnosing pea plant diseases from images. 
• A prototype smart monitoring system integrating image acquisition and automated analysis. 
• A decision-support tool for improving disease management and reducing crop loss. 
• A reproducible pipeline that can be extended to other crops.

Significance:

This project contributes to precision agriculture by reducing reliance on manual scouting and enabling early disease detection. It aligns with sustainable farming practices by optimizing resource use and minimizing unnecessary pesticide application. The integration of AI and IoT technologies can enhance crop productivity and support data-driven agricultural management.

Tools and Technologies:

• Data Sources: Field images, public plant disease datasets 
• Programming: Python, PyTorch, OpenCV, pretrained CNN/ViT models 

Timeline:

• 15 months