3D Automatic Target Recognition for UAV Platforms

Unmanned Aerial Vehicles (UAVs) using a Light Detection and Ranging (LIDAR) sensor can afford high quality 3D Automatic Target Recognition in complex environments. Considering the progress of object recognition algorithms in the computer vision domain, this paper evaluates current local based 3D descriptors on highly credible simulated military UAV reconnaissance scenarios. Trials involve a single and a multi template scheme while the UAV is flying at various flight parameters, the UAV - target distance is concerned and in parallel atmospheric perturbations are taken into account.

I. Introduction

Extending Automatic Target Recognition (ATR) from the 2D to the 3D domain offers numerous advantages. Indicatively, features (data) extracted from the 3D domain are less affected by illumination variation and target pose changes [1] and describe the underlying structure of an object. In addition, the Light Detection and Ranging's (LIDAR) short operating wavelength provides high-resolution data and the capability to acquire details of the target reinforcing recognition applications. With respect to Unmanned Aerial Vehicles (UAVs) by incorporating a LIDAR sensor 3D ATR can improve reconnaissance against camouflage and deception techniques.