It is estimated that a minimum of 5.25 trillion plastic objects weighing 268,940 tons are found in the world’s oceans [1]. Techniques such as ship observations, net trawling and water filtration help estimate the amount of plastic debris at the local scale [2]. The main goal of this research is to devise an automatic method for detecting and classifying marine debris. To this end, we are using two machine learning methods: A. Bag of Features B. CNN using Bottleneck to create classifiers able to classify images of marine debris in respective categories.

Citation: Kylili Kyriaki, Artusi Alessandro, Kyriakides Ioannis, & Hadjistassou Constantinos. (2018). Tracking and identifying floating marine debris. Sixth International Marine Debris Conference, San Diego California USA, 12-16 March 2018.