Application of Artificial Intelligence and Computer Vision for Measuring and Counting Oysters

Area of Science:

• Aquaculture technology
• Computer vision in agriculture
• Artificial intelligence applications

Background:

• Manual oyster size measurement is time-consuming and prone to variability.
• Accurate and efficient oyster size and count data are crucial for oyster farm management.

Purpose of the Study:

• To develop and validate an automated methodology for oyster counting and size estimation using artificial intelligence (AI) and computer vision (CV).
• To significantly reduce the time and improve the accuracy of oyster size and count measurements in aquaculture settings.

Main Methods:

• Image analysis employing DBScan clustering, artificial neural networks (ANNs), and random forest classification.
• Automated extraction of oyster length and width data from digital images.
• Comparison of automated measurements against manual methods.

Main Results:

• The AI-driven methodology achieved an 86.7-fold increase in speed for oyster length and width measurements compared to manual methods.
• The system demonstrated feasibility for automated oyster counting, with minor discrepancies in two out of ten images.
• Successful automatic classification and size estimation of oysters from images.

Conclusions:

• The proposed AI and CV methodology offers a feasible and highly efficient solution for automated oyster size measurement and counting in oyster farms.
• This technology has the potential to revolutionize oyster farming practices by improving operational efficiency and data accuracy.