Water Bath Scallop Shucking System Based on Doneness Detection
Guoliang Yang1,2, Xiangnian Shang1,2, Kai Cheng1,2
1Faculty of Information Science and Engineering, Ocean University of China, Qingdao 266100, China.
Sensors (Basel, Switzerland)
|July 28, 2026
Summary
This study introduces an automated scallop shucking system using visual feedback and temperature control. The system features an advanced AI model for real-time doneness detection, significantly improving shucking efficiency and scallop quality.
Area of Science:
- Computer Vision
- Artificial Intelligence
- Seafood Processing Technology
Background:
- Manual scallop shucking presents challenges in labor intensity and product quality consistency.
- Automated solutions are needed to improve efficiency and standardization in the seafood industry.
Purpose of the Study:
- To develop an intelligent scallop shucking system utilizing visual feedback and temperature control.
- To create a robust scallop doneness detection model for real-time assessment.
Main Methods:
- Systematic experiments to define baseline shucking parameters.
- Development of the SDD-RT-DETR model with custom modules (HierarchicalRepBlock, AIFI-EDFFN, EfficientBalanceFusion, Converse2DC3) and Wise-DIoU loss.
- Integration of the model into a water bath shucking system with feedback temperature control.
Main Results:
- The SDD-RT-DETR model achieved 95.5% accuracy, 93.6% recall, and 96.1% mAP50, outperforming the baseline RT-DETR.
- Reduced computational cost by 18.9% and parameters by 14.1% compared to the baseline.
- The integrated system demonstrated a 96.6% shucking rate and 88.5% properly cooked rate.
Conclusions:
- The developed AI model offers accurate, real-time scallop doneness detection, addressing a gap in computer vision applications.
- The automated shucking system enhances scallop quality and processing efficiency.
- This research provides a practical solution for the intelligent upgrade of the seafood processing industry.
