Fish Identification in Underwater Video with Deep Convolutional Neural Network: SNUMedinfo at LifeCLEF Fish task 2015

This paper describes our participation at the LifeCLEF Fish task 2015. The task is about video-based fish identification. Firstly, we applied foreground detection method with selective search to extract candidate fish object window. Then deep convolutional neural network is used to classify fish species per win- dow. Classification results are post-processed to produce final identification out- put. Experimental results showed effective performance in spite of challenging task condition. Our approach achieved best performance in this task.