Tracking Multiple Insects Using Multilayer Feed-Forward Networks

In the present study, attempts are made to capture and track coconut black headed caterpillar, Opisina arenosella and its parasitoid, Goniozus nephantidis with respect to their path and orientation. We devised an automatic tracking system using Artificial Neural Network for tracking both insects. The tracking system is based on the extracted features of the insects. Using the geometry of the image we have proposed a method to separate the two insects when they are joined in the image in the course of their motion. A fixed multilayer feed-forward backpropagation network (FMFBPN) was employed to solve the correspondence problem between frames. After establishing correspondence, the traced paths are plotted and length of the path of each insect is computed.