Image stitching techniques for an Intelligent Portable Aerial Surveillance System

摘要

Intelligent Portable Aerial Surveillance System (IPASS) is aimed at developing a low-cost, light-weight unmanned aerial vehicle that can provide sufficient battlefield intelligence for individual troops. IPASS consists of a ground station and an aircraft, which can take images high above the ground. To get a wider view of the ground, three images are taken simutaneously and then get stitched into a panorama. This paper first briefly introduces the system design of the project. Then it focuses on the processing of aerial images acquired from the three onboard cameras. The paper describes both the feature-based image stitching method and the infini-tehomography-based method for this task. Compared to the popular feature-based method, the second one can greatly reduce computation time with very little compromise in the quality of the panorama, which makes real-time video display of the surroundings on the ground station possible. This paper provides quantitative comparison of these two image stitching techniques to evaluate the performance in acquiring aerial surveillance information from multiple vision.