Feature-Based Target Recognition in Infrared Images for Future Unmanned Aerial Vehicles

This paper considers the task of object recognition for the possible introduction of target discrimination capabilities in unmanned aircraft. We are concerned with a naval scenario, and attempt to identify automatically four different ships in simulated infrared imagery, using a popular feature extractor and object recognition method: the Scale Invariant Feature Transform (SIFT). Features generated in test images are matched to features in a model database of labelled images using a modified minimum Euclidean distance metric. Pose clustering via the Generalised Hough Transform is also used to reduce outlying feature matches. Particular attention is paid to the robustness of the identification process to variations in scene illumination and the inclusion of Gaussian noise.