Plankton Image Classification using Convolutional Neural Networks

The study of plankton distribution is an important tool used for assessing the changes to marine ecosystem. Having a robust automated system for classification of plankton images plays an important role in advancing marine biology research. The images used in this study come from the SIPPER system. The challenges with SIPPER’s plankton image dataset are the high degree of similarities between different classes, high variability within the same class, partial occlusion, and noise. Also, traditional computer vision techniques require tedious work to find suitable features to represent plankton. To overcome those issues, we propose the use of convolutional neural networks. Results of the experiments on SIPPER dataset show improvement in classification accuracy in comparison to other state of the art approaches. Another major advantage of our approach is the scalability for classification of new classes without the need for feature engineering. Keywords—plankton images, SIPPER system, convolutional neural networks, image search.