DEEP LEARNING POWERED PESTS AND DISEASES DETECTION IN MAIZE

This project introduces a robust system for identifying pests and diseases in maize plants, offering tailored remedies for enhanced plant health.

Comment

Inspiration

The inspiration for this project stems from the critical role agriculture plays in Kenya's economy and the challenges posed by plant diseases and pests. The need for a more accurate and efficient method of detection led to the exploration of AI and image recognition techniques.

What it does

The developed system utilizes AI, particularly a Convolutional Neural Network (CNN), to analyze plant leaves through image recognition. It identifies pests or diseases affecting the plants, enabling timely interventions. Additionally, an IoT approach with an Nvidia Jetson allows real-time image capture and continuous monitoring for early detection.

How we built it

We built the system by training a CNN with a diverse dataset of labeled images, encompassing healthy leaves and leaves affected by various pests and diseases. Incorporating an extensive database of remedies and treatments, the system provides tailored recommendations based on identified issues. The IoT component uses an Nvidia Jetson with a camera module for efficient real-time analysis.

Challenges we ran into

Developing this AI-based system posed challenges in dataset curation, model training, and integrating the IoT component. Ensuring the system's accuracy and adaptability to diverse conditions required overcoming hurdles in both technical and practical aspects.

Accomplishments that we're proud of

We take pride in successfully implementing a robust AI system capable of accurately identifying plant diseases and pests. The integration of an IoT approach for continuous monitoring represents a significant achievement in enhancing agricultural practices.

What we learned

Through this project, we gained valuable insights into the complexities of training AI models for agricultural applications, the importance of diverse datasets, and the challenges associated with real-time monitoring using IoT devices. The interdisciplinary nature of the project enriched our understanding of both AI and agriculture.

What's next for DEEP LEARNING POWERED PESTS AND DISEASES DETECTION IN MAIZE

The next steps involve refining the system's accuracy, expanding the dataset for greater inclusivity, and exploring opportunities for scalability. Additionally, we aim to collaborate with agricultural stakeholders for real-world testing and implementation, further validating the system's effectiveness in transforming agricultural practices in Kenya.

Built With

Share this project:

Updates