Unsupervised classification and clustering of image features for vehicle detection in large scale aerial images

This paper presents a set of algorithms for vehicle detection in large scale aerial images. Vehicles are detected based on geometric and radiometric features, extracted within a multiresolution linear Gaussian scale-space. The image features, described by their local structures, are classified using support vector machines. Classified features are then clustered by an unsupervised affine propagation clustering algorithm, within a feature-level fusion scheme. Subcomponent of vehicles' body parts are aggregate together with respect to shared spatial relations and based on constraints on the orientation of detected vehicles. Experimental results using large scale aerial imagery demonstrate the efficient and robustness of the proposed algorithms for the detection of vehicles in an urban environment.