An Experimental Study on Estimating the Quantity of Fish in Cages Based on Image Sonar

Abstract

To address the highly demanding assessment of the quantity of fish in cages, a method for estimating the fish quantity in cages based on image sonar is proposed. In this method, forward-looking image sonar is employed for continuous detection in cages, and the YOLO target detection model with attention mechanism as well as a BP neural network are combined to achieve a real-time automatic estimation of fish quantity in cages. A quantitative experiment was conducted in the South China Sea to render a database for training the YOLO model and neural network. The experimental results show that the average detection accuracy mAP50 of the improved YOLOv8 is 3.81% higher than that of the original algorithm. The accuracy of the neural network in fitting the fish quantity reaches 84.63%, which is 0.72% better than cubic polynomial fitting. In conclusion, the accurate assessment of the fish quantity in cages contributes to the scientific and intelligent management of aquaculture and the rational formulation of feeding and fishing plans.