Car Detection in Low Resolution Aerial Images Tao Zhao Ram Nevatia University of Southern California Institute for Robotics and Intelligent Systems Los Angeles CA 90089-0273 taozhao nevatia @iris.usc.edu Abstract We present a system to detect passenger cars in aerial images along the road directions where cars appear as small objects. We pose this as a 3D object recognition problem to account for the variation in viewpoint and the shadow. We started from psychological tests to find important features for human detection of cars. Based on these observations, we selected the boundary of the car body, the boundary of the front windshield, and the shadow as the features. Some of these features are affected by the intensity of the car and whether or not there is a shadow along it. This information is represented in the structure of the Bayesian network that we use to integrate all features. Experiments show very promising results even on some very challenging images. Keywords Car Detection, Object Detection, Multi-Cue Integration, Bayesian Network, Aerial Image Analysis 1 Introduction Vehicle detection in aerial images has important civilian and military uses, such as traffic surveillance, both for traffic information system or to gather traffic statistics for urban planning. It can also produce strong evidence for road detection [11]. It also provides a good test domain for methods of object detection in difficult situation that require integration of multiple cues. The aerial images we used are grayscale images taken mostly from a vertical 1 This research was supported in part by a subgrant from MURI grand no. F49620-95-1-0457 from the US Army Research Office awarded to Purdue University.