An Underwater Image Enhancement Robotic System for Salmon Eggs Monitoring
CWSF · 2026 Agriculture, Fisheries & Food Bronze Medal
Overview
Pacific salmon stocks have great economic and cultural significance to Canadians. In recent years, many salmon species have been listed as at risk in salmon conservation assessments. An underwater robot-based monitoring system can provide a low-cost, non-invasive method for conducting salmon egg surveys, thereby helping researchers understand the hatching success of salmon at an early stage of their life cycle. We designed and built an underwater hexapod with a set of flexible fins that moves in wave-like motion to photograph salmon eggs in streams. We implemented several versions of the You Only Look Once algorithms to assess whether enhancing image quality of salmon eggs would count the number of eggs more accurately, especially in dark, murky waters or when eggs are piled on top of each other compared to counting using the original image. We believe our system can generate data to support stock assessment and management for salmon conservation.
Awards (2)
- Bronze Medal
- Selected for CWSF 2026
Competition history
- CWSF 2026
Related projects
ISEF · 2024
VannameiVision: A Novel End-to-End Pipeline for Stock Quality Assessment With a Case Study in Shrimp Post-larvae Screening From Robust Data Acquisition to Autonomous Segmentation and Systematic Characterization of Count, Size, and Vitality
ISEF · 2021
Protecting Critical Aquatic Ecology with Deep Learning
ISEF · 2022
WAL-SEA: Development of a Homebuilt, Multifunctional Remotely Operated Vehicle for the Study of the Near-Shore Ocean Ecosystems
ISEF · 2026
Autonomous Underwater Vehicle (AUV) for Real-Time 3D Microplastic Concentration and Toxicity Mapping via Deep Learning and Integrated Microscopy
Closest projects by meaning, across every fair and year in the corpus.