Fast detection of overlapped areas in smart camera network

Finding overlapped areas among images is a fundamental problem while building a multi-camera network. In a cooperative vision task, cameras must determine their relative positions with respect to each other. In this paper, a fast view matching based method is proposed to detect overlapped areas for un-calibrated camera network. We accomplish view matching by detecting interesting points matched with their local image descriptor. The detection of interesting points is based on the SURF algorithm, which is suitable for real time processing. The local image descriptor matching is built on the local binary pattern (LBP) histogram, which is robust for illuminating changes and image rotations. We have tested the algorithm in an actual embedded smart camera network. The performance is shown to be acceptable.