Classification of Red Mullet, Bluefish and Haddock Caught in the Black Sea by "Single Shot Multibox Detection"

Red Mullet, Haddock and Bluefish are usually caught at the same time on Turkey's Black Sea coast. Red Mullet, Haddock and Bluefish caught simultaneously are manually classified by the employees. This is a very arduous task due to the cold winter months in the Black Sea region. For this reason, automatic fish sorting machines are needed. In this study, Red Mullet, Haddock and Bluefish are classified with convolutional network. The advanced version of convolutional network: SSD (Single Shot Multi-Box Detection) method is used for classification of defined fishes. The network models of Red Mullet, Haddock and Bluefish is developed with TensorFlow framework. The model developed in the study will be used in the fish classification machine project that automatically classifies Red Mullet, Haddock and Bluefish.