Browsing by Subject "cameras"

Now showing 1 - 3 of 3
  • Thumbnail Image
    Item
    Predict Vehicle Collision by TTC From Motion Using a Single Video Camera
    (IEEE, 2018-05) Kilicarslan, Mehmet; Zheng, Jiang Yu; Computer and Information Science, School of Science
    The objective of this paper is the instantaneous computation of time-to-collision (TTC) for potential collision only from the motion information captured with a vehicle borne camera. The contribution is the detection of dangerous events and degree directly from motion divergence in the driving video, which is also a clue used by human drivers. Both horizontal and vertical motion divergence are analyzed simultaneously in several collision sensitive zones. The video data are condensed to the motion profiles both horizontally and vertically in the lower half of the video to show motion trajectories directly as edge traces. Stable motion traces of linear feature components are obtained through filtering in the motion profiles. As a result, this avoids object recognition and sophisticated depth sensing in prior. The fine velocity computation yields reasonable TTC accuracy so that a video camera can achieve collision avoidance alone from the size changes of visual patterns. We have tested the algorithm for various roads, environments, and traffic, and shown results by visualization in the motion profiles for overall evaluation.
  • Thumbnail Image
    Item
    Secret key distribution leveraging color shift over visible light channel
    (IEEE, 2017-10) Liu, Hongbo; Liu, Bo; Shi, Cong; Chen, Yingying; Computer and Information Science, School of Science
    Given the widely adoption of screen and camera in many electronic devices, the visible light communication (VLC) over screen-to-camera channel emerges as a novel short range communication technique in recent years. Active research explores various ways to convey messages over screen-camera channel, such as barcode and unobtrusive optical pattern. However, with the prevalence of LED screens of wide viewing angles and mobile devices equipped with high standard cameras, the threat of information leakage over screen-to-camera channel becomes in-negligible. Few studies have discussed how to ensure the security of data transmission over screen-to-camera channel. In this paper, we propose a secret key distribution system leveraging the unique color shift property over visible light channel. To facilitate such design, we develop a practical secret key matching based method to map the secret key into gridded optical patterns on screen, which can only be correctly recognized by the legitimate user through an accessible region and allow regular data stream transmission through valid grids. The proposed system is prototyped with off-the-shelf devices and validated under various experimental scenarios. The results show that our system can achieve high bit-decoding accuracy for the legitimate users while maintaining comparable data throughput as regular unobtrusive VLC systems with very low recovery accuracy of the encrypted data for the attackers.
  • Thumbnail Image
    Item
    Vision-Based Target Tracking and Autonomous Landing of a Quadrotor on a Ground Vehicle
    (IEEE, 2017-05) Hoang, Tru; Bayasgalan, Enkhmurun; Wang, Ziyin; Tsechpenakis, Gavriil; Panagou, Dimitra; Computer and Information Science, School of Science
    This paper addresses vision-based tracking and landing of a micro-aerial vehicle (MAV) on a ground vehicle (GV). The camera onboard the MAV is mounted so that the optical axis is aligned with the downward-facing axis of the body-fixed frame. A novel supervised learning vision algorithm is proposed as the method to detect the ground vehicle in the image frame. A feedback linearization technique is developed for the MAV to fly over and track the GV so that visibility with the tracked target is maintained with certain guarantees. The efficacy of the visual detection algorithm, and of the tracking and landing controller is demonstrated in simulations and experiments with static and mobile GV.