Research advances on fish feeding behavior recognition and intensity quantification methods in aquaculture

2025-02-21

AI Summary

This study addresses the low accuracy and poor robustness of fish feeding behavior recognition and quantification in aquaculture. We systematically review and empirically evaluate single-modal approaches—including computer vision, underwater acoustics, and physical sensors—as well as multimodal fusion techniques. For the first time, we conduct a cross-modal comparative analysis of feeding intensity quantification, identifying performance limits and practical deployment bottlenecks under challenging conditions such as low illumination, high turbidity, and multi-fish occlusion. We propose an environment-aware technology selection framework integrating object detection, action recognition, acoustic time-frequency analysis, cross-modal feature alignment, and adaptive weighted fusion. The framework delivers a deployable methodology and technical roadmap for intelligent feeding systems, significantly improving both health monitoring accuracy and feeding decision efficiency in real-world aquaculture settings.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data

Abstract

As a key part of aquaculture management, fish feeding behavior recognition and intensity quantification has been a hot area of great concern to researchers, and it plays a crucial role in monitoring fish health, guiding baiting work and improving aquaculture efficiency. In order to better carry out the related work in the future, this paper firstly reviews the research advances of fish feeding behavior recognition and intensity quantification methods based on computer vision, acoustics and sensors in a single modality. Then the application of the current emerging multimodal fusion in fish feeding behavior recognition and intensity quantification methods is expounded. Finally, the advantages and disadvantages of various techniques are compared and analyzed, and the future research directions are envisioned.

Problem

• Develop fish feeding behavior recognition methods
• Quantify feeding intensity for health monitoring
• Explore multimodal fusion for aquaculture efficiency

Innovation

• Computer vision for behavior analysis
• Acoustic methods for intensity quantification
• Multimodal fusion enhances recognition accuracy