Abstract:Feed has the greatest impact on the carbon footprint of the aquaculture, and also determines the water quality in aquaculture to a great extent. Making appropriate feeding control strategies is one of the most effective ways to promote cleaner production as well as fish welfare in aquaculture. Reliable and accurate fish appetite grading especially based on time-limited data is a prerequisite for achieving high-precision and reasonable feeding control in practical production. To date, however, few efforts have been done on this challenge. For these, regarding Micropterus salmoides as the experimental fish, a novel and practical method, based on a modified kinetic energy feature-based graph convolutional network (GCN), was developed in this study. First, graphs were constructed based on the extracted modified kinetic energy features and their temporal correlation. Then, with the help of a series of the convolution and global pooling operations, a GCN model was customized based on the constructed graphs. Following this, the customized GCN model was enriched by the self-attention pooling mechanism and customized network structure. Results show that the proposed GCN-based approach outperforms other typical state-of-the-art methods in fish appetite grading, and the grading accuracy obtained here could be 98.60% using only the first 4.2 seconds as well as the first 8.3 seconds of input data, which is not much different from that (98.89%) using full-length (25 second-long) input data. What’s more, compared to the recurrent neural network (RNN)-based method which performance is closest to our method, the space complexity of the proposed approach here can better satisfy the requirements of real aquaculture, in which the quantity of the trainable parameters here is only 6.4% ~ 31.8% of the RNN-based method. In summary, the proposed modified kinetic energy feature-based GCN approach is favorable for the appetite grading of fish like Micropterus salmoides with time-limited data, which is a promising approach in dealing with feeding control tasks and alleviating the water environmental burden in aquaculture.

GCVC: Graph Convolution Vector Distribution Calibration for Fish Group Activity Recognition

Modified Kinetic Energy Feature-Based Graph Convolutional Network for Fish Appetite Grading Using Time-Limited Data in Aquaculture

Modified Motion Influence Map and Recurrent Neural Network-Based Monitoring of the Local Unusual Behaviors for Fish School in Intensive Aquaculture.

Multi-class Fish Stock Statistics Technology Based on Object Classification and Tracking Algorithm

Semi-supervised learning-based live fish identification in aquaculture using modified deep convolutional generative adversarial networks

Fish Detection and Classification Based on Improved ViT

An Automatic Detection and Statistical Method for Underwater Fish Based on Foreground Region Convolution Network (FR-CNN)

Vision-based dual network using spatial-temporal geometric features for effective resolution of fish behavior recognition with fish overlap

FCFormer: fish density estimation and counting in recirculating aquaculture system

Dynamical Attention Hypergraph Convolutional Network for Group Activity Recognition

Boosting fish counting in sonar images with global attention and point supervision

Automatic Recognition of Fish Behavior with a Fusion of RGB and Optical Flow Data Based on Deep Learning

DC3D: A Video Action Recognition Network Based on Dense Connection

HiGCIN: Hierarchical Graph-Based Cross Inference Network for Group Activity Recognition

A recurrent network based on active learning for the assessment of fish feeding status

GT-CAM: Game Theory Based Class Activation Map for GCN

Coordinate-Aware Mask R-CNN with Group Normalization: A underwater marine animal instance segmentation framework

Robust Tracking of Fish Schools Using CNN for Head Identification

Multi-dimensional convolution transformer for group activity recognition

Fish detection based on Gather-and-Distribute mechanism Multi-scale feature fusion network and Structural Re-parameterization method

Educational tool for hospital-based training in family medicine.