Medaka Observation Based on Rotation Invariant Features

Recently, image processing is widely used in research on the habits and characteristics creatures. For example, the medaka (Oryzias laptipes) which breeds quickly is popular used in various experiments. In particular, observing the behavior of the medaka is important to understanding the ecology of their characteristics. In this paper, we proposed the tracking method based on image processing for a medaka observation of its behavior. Therefore, we describe our proposed tracking method which combine the estimated position and the orientation code matching based on rotation invariant feature. Through experimentation, we examined the effectiveness of our proposed method.