QoI-Aware Camera Network-as-a-Service for Social Behavior Analysis

Abstract

The frames captured by the camera should be of sufficient quality, generally measured in pixel density on a target (pixels per foot or pixels per meter (ppm)) to derive information suitable for social behavior analysis. Achieving 24x7 high pixel density coverage incurs a large number of cameras and exorbitant network bandwidth for large scale deployments. In this paper, we introduce additional features to the Camera Network-as-a-Service (CNaaS) concept proposed by Misra et al. to address two aspects pertinent to services related to social behavioral analysis. We present a Quality of Information-Aware (QoI-Aware) CNaaS to address the two aspects. (1) To allow the reuse of cameras which are already available in a locality to minimize the number of additional cameras required to provide the CNaaS service. (2) Inclusion and exclusion of camera node(s) into the CNaaS platform while ensuring fair opportunities to the Camera Network Owners (CNOs) which provide the same Quality of Information (QoI). The simulation results indicate that the proposed scheme is fair to nodes providing the same QoI, reduces the energy consumption, and excludes the cameras which offer lesser QoI. This will lead to an improvement in the usage of network bandwidth, and the profit of the Camera Network Service Provider (CNSP).