Scalable Fish Segmentation And Species Classification: A Case Study On Turkey’s Aquatic Biodiversity

This study aims to develop an accurate classification model to identify fish species based on unique physical characteristics such as texture, color, and shape. A dataset containing 1,000 images per fish species was collected from a supermarket fish counter in Izmir, Turkey, encompassing eight distinct species. Machine learning classifiers, including K-Nearest Neighbors (KNN) and Support Vector Machine (SVM), were employed to analyze and categorize the species effectively. Both classifiers demonstrated exceptional performance, achieving 100% accuracy in distinguishing the fish species. These results underscore the ability of KNN and SVM to reliably classify fish based on specific environmental conditions, such as the supermarket display setting. The findings suggest that these classification models can be instrumental in applications requiring precise species identification, such as fisheries management, biodiversity monitoring, and supermarket inventory systems. This research highlights the potential of machine learning in addressing real-world challenges in aquatic biology and food supply chain management.