Intelligent Automation of Penaeid Shrimp Feeding

Photo of author

By Milthon Lujan

Overall process flow of the proposed method. Source: Yu-Kai et al., (2023), Connection Science.
Overall process flow of the proposed method. Source: Yu-Kai et al., (2023), Connection Science.

The global shrimp farming industry is currently undergoing a remarkable transformation, characterized by rapid growth and innovation. One of the key drivers of this evolution is the adoption of cutting-edge automation techniques, which offer substantial benefits in terms of cost reduction and decreased labor demands. An exemplary instance of this transformation is the automation of shrimp feeding.

Researchers from Providence University, National Taiwan University, and Clarkson University propose an intelligent system for shrimp farming that includes shrimp detection, measurement of shrimp length, shrimp quantity, and two methods for determining the degree of stomach fullness in the shrimp.

The scientists delved into the realm of intelligent shrimp farming and explored how the integration of AR-YOLOv5, a novel technology, is revolutionizing the industry. This innovative approach not only enhances shrimp growth but also contributes to the environmental sustainability of the shrimp farming industry and optimizes the shrimp feeding process.

Precision in Shrimp Detection and Measurement

One of the fundamental components of this automation revolution is the precise detection and measurement of shrimp. This involves accurately estimating shrimp length and quantity, which is crucial for efficient farm management. The introduction of AR-YOLOv5 (Angular Rotation YOLOv5) changes the game in this regard. Researchers have incorporated it into the system to significantly improve the performance and efficiency of shrimp detection and measurement.

Shrimp Detection and Measurement with AR-YOLOv5

The AR-YOLOv5 system, which stands for Angular Rotation YOLOv5, is an advanced deep learning model designed to detect and measure objects with unparalleled accuracy. In the context of shrimp farming, this technology is harnessed for precise shrimp detection and the estimation of their length and quantity.

Experiments conducted in a real shrimp farming environment produced remarkable results. When using AR-YOLOv5, the system achieved a precision rate of 97.70%, a recall rate of 91.42%, a mean average precision of 94.46%, and an F1-score of 95.42% in terms of detection performance. These metrics clearly highlight the potential of AR-YOLOv5 to revolutionize shrimp farming through precise and efficient shrimp detection and measurement.

Optimizing Feeding with Stomach Fullness Measurement

Effective feeding management is crucial in shrimp farming to reduce food waste and water contamination. Feeding based on the degree of stomach fullness can significantly enhance the sustainability of shrimp farming practices. During feeding, the shrimp’s digestive tract becomes full, leading to an increase in the stomach repletion index, which represents the degree of stomach fullness.

“The stomach repletion index is used to measure the stomach fullness levels of shrimp, where 0% indicates empty intestines and 100% represents fully filled intestines,” cite the researchers.

The proposed intelligent shrimp farming system has introduced not one but two methods for determining the degree of stomach fullness.

The system excels in stomach fullness measurement. It achieves an impressive accuracy rate of 88.8%, a precision rate of 91.7%, a recall rate of 90.9%, and an F1-score of 91.3% when determining stomach fullness in real shrimp farming environments. These results underscore the practical significance of the proposed system in improving feeding practices, reducing food waste, and minimizing water contamination.

The Promise of Intelligent Shrimp Farming

The adoption of an intelligent shrimp farming system that combines precise shrimp detection and measurement with accurate stomach fullness determination offers numerous advantages. It not only improves the operational efficiency of shrimp farms but also promotes environmental sustainability by minimizing resource wastage.

However, it is important to note that the researchers identified the following limitations:

  • When water quality is severely turbid due to shrimp feed and excrement accumulation, visibility can be adversely affected.
  • As shrimp grow, their shells become thicker and darker, making their stomach and digestive tract barely visible.
  • Researchers only measured the size of shrimp that were fully visible within the camera’s scope.


In conclusion, the shrimp farming industry is at the forefront of innovation, driven by the adoption of automation techniques and cutting-edge technologies like AR-YOLOv5. The impressive results achieved in terms of shrimp detection, measurement, and stomach fullness determination clearly indicate that intelligent automation has the potential to revolutionize the industry. As shrimp farming continues to grow, the integration of such technologies becomes not just a choice but a necessity for the industry’s sustainability and prosperity.

This research was partially funded by the Higher Education Sprout Project, Ministry of Education, at the University Advancement headquarters of the National Cheng Kung University (NCKU).

Tien-Hsiung Weng
Department of Computer Science and Information Engineering
Providence University
Taichung, Taiwan, ROC
Email: thweng@gm.pu.edu.tw

Reference (open access)
Yu-Kai Lee, Bo-Yi Lin, Tien-Hsiung Weng, Chien-Kang Huang, Chen Liu, Chih-Chin Liu, Shih-Shun Lin & Han-Ching Wang (2023) Counting and measuring the size and stomach fullness levels for an intelligent shrimp farming system, Connection Science, 35:1, DOI: 10.1080/09540091.2023.2268878

Leave a Comment