Deep Learning-Based Image Recognition for Soft-Shell Shrimp Classification
With the integration of information technology into aquaculture, production has become more stable and continues to grow annually. As consumer demand for high-quality aquatic products rises, freshness and appearance integrity are key concerns. In shrimp-based processed foods, freshness declines rapidly post-harvest, and soft-shell shrimp often suffer from head-body separation after cooking or freezing, affecting product appearance and consumer perception. To address these issues, this study leverages deep learning-based image recognition for automated classification of white shrimp immediately after harvest. A convolutional neural network (CNN) model replaces manual sorting, enhancing classification accuracy, efficiency, and consistency. By reducing processing time, this technology helps maintain freshness and ensures that shrimp transportation businesses meet customer demands more effectively.