Enhanced Shrimp Disease Classification through YOLO and Data Augmentation

The global aquaculture industry, crucial for meeting seafood demand, faces significant challenges, notably disease outbreaks causing substantial financial losses, particularly in shrimp farming. This research introduces an innovative system to manage shrimp health in aquaculture ponds. The system employs a multi-faceted approach encompassing segmentation, object tracking algorithm SORT, and real-time notifications. Using the YOLOV8 segmentation algorithm, it accurately distinguishes healthy and infected shrimp. Additionally, an object tracking algorithm SORT enables continuous monitoring, precise quantification of shrimp populations, and informed disease management decisions, reducing financial losses from outbreaks. A unique feature is the system's real-time notification capability, providing shrimp farm owners with up-to-the-minute counts of infected shrimp, enabling prompt action to contain and treat infections, ultimately minimizing losses and enhancing overall farm management. In summary, this research offers an integrated solution utilizing advanced computer vision and tracking technologies to improve shrimp health monitoring in aquaculture, allowing farm owners to proactively manage ponds, mitigate disease outbreaks, and reduce financial losses.