부산대학교 도서관

Insect behavioral studies are time consuming but essential in pest management. Tracking different kinds of insects is an integral part of these studies. In this paper, we present a new approach to track multiple insects especially the larva which is elastomeric in nature. This study demonstrates the use of image processing techniques for segmenting insects, a detailed procedure for separating the head and tail sections from the larva trunk to minimize the errors due to deformation, invariant moment features to establish correspondence across frames to arrange insect parts in order. This study automates the host searching behavior of parasitoid species, which are used for biological control programs.