Monitoring Trichogramma Activities From Videos An adaptation of cognitive vision system to biology field

This paper describes a new adaptation of a cognitive vision system in biology field. Our goal is to monitor Trichogramma activities from image sequence. More precisely we aim at recognizing several scenarios of the parasitoid behavior when it parasites the eggs of the Mediterranean flour moth Ephestia kuehniella Zeller (crops-harmful butterfly). The main advantage of our current work is the use of accurate features for characterizing mobile objects shapes. We first establish different requirements to describe Trichogramma activities involved in the laboratory experiment context. We second explain the steps taken to adapt a general video understanding framework containing a scenario recognition algorithm to reach our objective.