OysterNet: Enhanced Oyster Detection Using Simulation

TL;DR

OysterNet leverages simulated oyster images based on geometric modeling to significantly improve underwater oyster detection accuracy with limited real data, aiding ecological preservation efforts.

Contribution

The paper introduces a novel simulation-based data augmentation method for oyster detection, enhancing performance and reducing reliance on extensive real datasets.

Findings

Up to 35.1% performance boost using synthetic data

Improved state-of-the-art accuracy by 12.7%

Demonstrates geometric modeling benefits in limited-data scenarios

Abstract

Oysters play a pivotal role in the bay living ecosystem and are considered the living filters for the ocean. In recent years, oyster reefs have undergone major devastation caused by commercial over-harvesting, requiring preservation to maintain ecological balance. The foundation of this preservation is to estimate the oyster density which requires accurate oyster detection. However, systems for accurate oyster detection require large datasets obtaining which is an expensive and labor-intensive task in underwater environments. To this end, we present a novel method to mathematically model oysters and render images of oysters in simulation to boost the detection performance with minimal real data. Utilizing our synthetic data along with real data for oyster detection, we obtain up to 35.1% boost in performance as compared to using only real data with our OysterNet network. We also improve the state-of-the-art by 12.7%. This shows that using underlying geometrical properties of objects can help to enhance recognition task accuracy on limited datasets successfully and we hope more researchers adopt such a strategy for hard-to-obtain datasets.