Underwater Fish Species Recognition Using Deep Learning Techniques

Abstract

Underwater fish species recognition has gained importance due to the emerging researches in marine science. Automating the fish species identification using technology would help the marine science to evolve further. Image classification tasks have seen a rise with the introduction of deep learning techniques. In this paper, we have proposed a hybrid Convolutional Neural Network (CNN) framework that uses CNN for feature extraction and Support Vector Machine (SVM) and K-Nearest Neighbour (k-NN) for classification. Both the proposed frameworks are tested on Fish4Knowledge dataset. Our experimental results show that our framework gives better results than most of the traditional as well as existing deep learning techniques.