Rahul Kavi

West Virginia University, MGW, West Virginia, United States

Are you Rahul Kavi?

Claim your profile

Publications (3)0.56 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: This article describes the use of wireless communication technology in an easy-to use ad hoc mode to address concerns of timely proximity warning and collision avoidance in surface mines and also describes the design of a cloud-based logging framework for long-term vehicular traffic analysis in mines. For timely warning about approaching vehicles at large distances (10–100 m), a GPS system is integrated with Wi-Fi (IEEE 802.11a/b/p) radios in an ad hoc mode, where information about approaching vehicles is known as soon as they come into range. A communication range test is performed in an actual surface mine setting to characterise the distances at which warning can be reliably received using each of the IEEE 802.11 family of radios. A zone-based proximity warning system is designed using low power IEEE 802.15.4 radios for detecting obstacles and vehicles at much smaller distances (<10 m), and marking them into zones around the vehicle. Both the proximity warning system and the Wi-Fi-based collision avoidance system were evaluated for feasibility at an operating surface coal mine in the southern United States. Finally, the design of a cloud-based logging framework is described and can be used for long-term data collection from GPS and other sensors.
    International Journal of Mining Reclamation and Environment 10/2015; 29(5):331-346. DOI:10.1080/17480930.2015.1086550 · 0.56 Impact Factor
  • Rahul Kavi · Vinod Kulathumani ·

    09/2013; 2(3):486-508. DOI:10.3390/jsan2030486
  • Source
    S. Ramagiri · R. Kavi · V. Kulathumani ·
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, we describe how information obtained from multiple views using a network of cameras can be effectively combined to yield a reliable and fast human action recognition system. We describe a score-based fusion technique for combining information from multiple cameras that can handle arbitrary orientation of the subject with respect to the cameras. Our fusion technique does not rely on a symmetric deployment of the cameras and does not require that camera network deployment configuration be preserved between training and testing phases. To classify human actions, we use motion information characterized by the spatio-temporal shape of a human silhouette over time. By relying on feature vectors that are relatively easy to compute, our technique lends itself to an efficient distributed implementation while maintaining a high frame capture rate. We evaluate the performance of our system under different camera densities and view availabilities. Finally, we demonstrate the performance of our system in an online setting where the camera network is used to identify human actions as they are being performed.
    Distributed Smart Cameras (ICDSC), 2011 Fifth ACM/IEEE International Conference on; 09/2011