Novel Embedded System Based Species Recognition System for Pest Control

Abstract

To develop a species recognition system for a resettable trap using novel species identification techniques. Classical pest control techniques are currently used to identify the pest population in New Zealand forests. The main interest of Department of Conservation (DOC) is to identify the pest before setting the traps and collect pest population information regarding certain types of pests. The main aim of this work is to design and develop a novel robust system which can be used to identify the animal in real time in different environmental conditions. We propose an image recognition technique based system to identify the pest. To be specific, identifying pests by body features and fur patterns color, using image processing techniques. For the test system, a GumstixOvero Fire COM (computer-on-module) with a Texas Instruments OMAP embedded platform is used to run the image processing algorithms on a Windows based CE operating system.