Bangladeshi crops leaf disease detection using YOLOv8

Area of Science:

• Agricultural Science
• Computer Vision
• Plant Pathology

Background:

• Leaf diseases pose a significant threat to crop yields and food security in Bangladesh.
• Major crops like rice, corn, wheat, potato, and tomato are highly vulnerable.
• There is a critical need for automated systems for early disease detection and management.

Purpose of the Study:

• To evaluate the efficacy of the YOLOv8 object detection model for automated leaf disease identification.
• To compare YOLOv8's performance against previous models in detecting diseases in key Bangladeshi crops.
• To enhance the efficiency of crop disease detection and management through advanced computer vision techniques.

Main Methods:

• A dataset of 2850 images across 19 classes (leaf disease types) was curated and annotated.
• The YOLOv8 (You Only Look Once) deep neural network framework was employed for training.
• Performance was evaluated using standard metrics including mean Average Precision (mAP) and F1 score.

Main Results:

• The YOLOv8 framework achieved a high mean Average Precision (mAP) of 98%.
• The system demonstrated a strong F1 score of 97% in disease identification.
• YOLOv8 successfully detected and categorized leaf diseases in the studied crops.

Conclusions:

• YOLOv8 shows significant potential for accurate and efficient automated crop disease management.
• This technology can contribute to improved food security and agricultural sustainability in Bangladesh.
• The study highlights the effectiveness of state-of-the-art computer vision for agricultural challenges.