Water Quality Monitoring Technologies

BettaAI: a machine-learning object detection model for the quantification of aggressive displays in the Siamese fighting fish Betta splendens

BettaAI is a deep learning model that accurately detects and quantifies aggressive displays in Siamese fighting fish using video analysis.

From video to encounter histories: individual identification and machine vision for salmonid capture-recapture monitoring

This paper develops a machine-vision workflow to convert video footage of salmonids into individual encounter histories for capture-recapture analysis, achieving high accuracy in fish detection and re-identification.

A Low-Resource Machine-Learning Framework for Cold-Stress Early Warning in Aquaculture Nursery Ponds Using Manual Temperature Readings

The paper develops a low-resource machine learning framework using manual temperature readings to predict cold stress in aquaculture ponds, achieving effective early warnings with XGBoost models.

Estimation of the curved body length of tuna larvae from microscope images using a zero-shot model and image processing techniques

This paper develops an image-processing method to estimate curved body length of tuna larvae from microscope images without training, achieving substantial agreement with manual measurements for larvae ≤5.0 mm.

Automated morphometry and weight prediction of juvenile Chinook Salmon leveraging open-source deep learning models

This paper develops an automated system using deep learning to estimate morphometric measurements and predict weight of juvenile Chinook Salmon from 2D images with high accuracy.

FishWIO: a labeled image dataset of Western Indian Ocean reef fishes for training and testing classification algorithms

FishProx: Proximate conspecific interaction of Atlantic salmon (Salmo salar) for behavioural analysis through instance segmentation masks

FishProx uses instance segmentation to analyze fish behavior in tanks, calculating welfare metrics through anomaly scores without requiring labeled data.

Using supervised machine learning to quantify cleaning behaviour

This paper develops a semi-automated system using DeepLabCut and a classification algorithm to quantify cleaning behaviour in cleaner wrasse and powder blue tang, achieving 90% accuracy and reducing manual annotation by 75%.

A real-time detection and non-destructive warning method for zebrafish body surface anomalies based on improved YOLOv8 framework

This paper introduces ESC-YOLOv8-seg, a real-time, non-destructive method for detecting zebrafish surface anomalies with high accuracy and speed.